Novel Mechanisms Regulating the Netrin-1 Receptor Deleted in Colorectal Cancer (DCC) During Cortical Axon Outgrowth

Novel mechanisms regulating the netrin-1 receptor deleted in colorectal cancer (DCC) during cortical axon outgrowth

Philippe M. Duquette

Degree of Doctorate of Philosophy

Department of Anatomy & Cell Biology

McGill University

Montreal, Quebec, Canada

November, 2018

A thesis submitted to McGill University in partial fulfillment of the requirements of the degree of Doctorate of Philosophy

©Copyright Philippe M. Duquette 2018

“The whole problem with the world is that fools and fanatics are always so certain of themselves and wiser people so full of doubts” Bertrand Russell

DEDICATION

I dedicate this thesis to the loving memory of my father, Richard Duquette - Papa, whose enthusiasm and interest in my research forced me to put the scientific jargon aside and practice my elevator pitch on a regular basis. I am forever grateful for the sacrifices you made so I wouldn’t have to. My accomplishments are as much yours as they are mine!

« Prend le temps de le faire comme il faut! »

iii

ACKNOWLEDGMENTS

First, I would like to express my gratitude to my Ph.D. supervisor, Dr. Nathalie Lamarche-Vane, who has been very supportive throughout my time in her laboratory. Nathalie’s trust has allowed me to grow and pave my own path. She was always willing to help, whether through discussion in her office or getting her hands dirty in the lab. Thank you for making my PhD that more fulfilling by sending me to a memorable microscopy course in Cold Spring Harbor. When confronted with the loss of several family members, in particular my father, she was always willing to help with concrete actions and went out of her way to help, however she could. She would offer her help not only as a mentor but as a friend and I am forever grateful to her. I couldn’t have asked for a more understanding supervisor. I acknowledge the Fonds de recherche du Québec-Santé, the RI-MUHC, McGill University and the Department of Anatomy and Cell Biology for their financial support. I also thank those who donated expertise, time and spirit towards the completion of this thesis. Thank you to the members of my advisory committee, Drs. Alyson Fournier and Lisa Munter, for their scientific input and encouragement as well as my mentor, Dr. Elaine Davis. I also would like to thank Dr. Isabelle Rouiller for acting as my mentor and helping with the netrin-1 purification in her lab. I have had the privilege to benefit from the valuable expertise of many researchers. Special thanks to Judith Antoine-Bertrand, Dana Hamam, Veronica Atehortua, Zehra Khoja, Zarina D’Costa, Sadig Niftullayev, and Vanessa Looz for the science, the laughs and the support. Finally, I would like to thank my family. My sister, Mélanie, for printing all those articles at her work. I did read all of them and that acquired knowledge helped me throughout my studies. Her interest and positive attitude helped me push through difficult times. My mother, Mom, who would be proud of me no matter what passed on her thirst for knowledge and learning through her love of reading. The hours spent at the library and bookstores opened up a whole universe of knowledge ready to be picked up and I still can’t wait to pick up the next one. Thanks for your unconditional love. My girlfriend, Siri, who put up with my frustrations, bad moods, and crankiness. I’m surprised she’s still there! Thanks for everything and especially your wonderful cooking. I love you. Laslty, Charline Malichanh, for her energy, playfulness, and her curiosity. Thanks “Madame Pourquoi”.

Table of Contents DEDICATIONS...... iii

ACKNOWLEDGMENTS………………………………………………………………………iv

TABLE OF CONTENTS………………………………………………………………………...v

LIST OF FIGURES……………………………………………………………………………..ix

LIST OF ABREVIATIONS…………………………………………………………………....xii

ABTSRACT…………………………………………………………………………………..xvii

RÉSUMÉ………………………………………………………………………………………xix

ORIGINAL SCHOLARSHIP AND CONTRIBUTIONS TO KNOWLEDGE……………….xxi

CONTRIBUTIONS OF AUTHORS…………………………………………………………xxiii

Preface to chapter 1………………………………………………………………………………1

Chapter 1: Introduction and Literature Review

1.1: The Nervous System………………………………………………………………………...2

1.1.1: The Cell Types of the Nervous System

1.1.1.1: Neurons………………………………………………………………….3

1.1.1.2: Glial Cells……………………………………………………………….4

1.1.2: The Central Nervous System……………………………………………………...4

1.2: Cellular diversity in the cortex

1.2.1: Arealization………………………………………………………………………..5

1.2.2: Nomenclature……………………………………………………………………...6

1.2.3: Cortical neuron subtypes – projection neurons……………………………………7

1.2.4 : Cortical neuron subtypes – interneurons………………………………………….9

1.3: From Neural Induction to Cortical Assembly

1.3.1: Neural induction………………………………………………………………….10

1.3.2: Cortical assembly………………………………………………………………...11

1.4: Neuronal polarization, axon specification, outgrowth and guidance

1.4.1: Neuronal Polarization……………………………………………………………14

1.4.2: Mechanisms regulating neuronal polarization

1.4.2.1: TRIM46………………………………………………………………..16

1.4.2.2: PI3K/PIP3……………………………………………………………...17

1.4.2.3: BDNF → TrkB → PKA → LKB1 → SAD-A/B/MARK-2…………..19

1.4.2.4: Ras superfamily of small GTPases…………………………………….20

1.4.2.5: Ras homologous (Rho) GTPases………………………………………21

1.4.2.6: In vivo………………………………………………………………….22

1.4.3: Axon outgrowth………………………………………………………………….25

1.4.3.1: The growth cone………………………………………………………..26

1.4.3.2: Actin – the driver………………………………………………………28

1.4.3.3: Microtubules – more than just an engine………………………………31

1.4.3.4: Actin/Microtubule interaction………………………………………….32

1.4.3.5: Intermediate filaments (Neurofilaments)………………………………33

1.4.3.6: Ras GTPases…………………………………………………………...33

1.4.3.7: Rho GTPases…………………………………………………………...34

1.4.4: Axon guidance – the road signs………………………………………………….35

1.4.4.1: Classical guidance cues and commissural axon guidance……………..36

1.4.4.2: Morphogens and Growth factors in commissural axon guidance……...37

1.4.4.3: Classical guidance cues

1.4.4.3.1: Slit/Robo signaling at the roundabout………………………..39

1.4.4.3.2: Semaphorin/Plexin signaling………………………………....41

1.4.4.3.3: Ephrin/Eph signaling…………………………………………42

1.4.4.3.4: Netrin – the one who guides…………………………………43

1.4.4.3.5: Netrin – the one who guides locally………………………….45

1.4.5: DCC – the one who attracts

1.4.5.1: Structure………………………………………………………………..45

1.4.5.2: Signaling……………………………………………………………….47

1.4.5.3: Netrin/DCC and the cytoskeleton……………………………………...50

1.4.5.4: DCC in human axon guidance…………………………………………51

1.4.5.5: DCC in neurological and psychiatric diseases…………………………53

1.4.6: Receptor regulation

1.4.6.1: Regulated endocytosis/exocytosis……………………………………...53

1.4.6.2: Proteolysis……………………………………………………………...54

1.4.7: Netrin-mediated cortical axon guidance…………………………………………57

1.5: Calpain……………………………………………………………………………………..61

1.5.1: Calpain regulation………………………………………………………………..62

1.5.2: Calpain in axon outgrowth and guidance………………………………………...64

Rationale and objectives………………………………………………………………………..65

Preface to chapter 2……………………………………………………………………………..67

Chapter 2: p120RasGAP Contributes as a Scaffolding Protein to Regulate Netrin-1-Mediated Axon Outgrowth and Attraction in a GAP-dependent……………………..68

Preface to chapter 3……………………………………………………………………………..86

Chapter 3: A role for the calcium-activated protease calpain in the regulation of netrin-1/DCC-mediated cortical axon outgrowth and guidance……………………………….87

Preface to chapter 4……………………………………………………………………………103

Chapter 4: Investigating the role of DCC-Thr 1210 phosphorylation during cortical axon Outgrowth……………………………………………………………………………………..104

Chapter 5 – Discussion……………………………………………………………………….108

5.1: Major findings…………………………………………………………………………….108

vii

5.2: p120RasGAP mediates Ras and netrin/DCC signaling………………………...………....108

5.2.1: p120RasGAP’s GAP-independent function…………………………………….109

5.2.2: p120RasGAP and other GAPs………………………………………………….111

5.3: Calpain-1/2 in netrin/DCC signaling……………………………………………………..113

5.3.1: Netrin/DCC/Calpain in cortical axon outgrowth……………………………….114

5.3.2: Netrin/DCC/Calpain in cortical axon guidance………………………………...115

5.3.3: Netrin/DCC/Calpain and ERK1/2 activation…………………………………...118

5.3.4: Netrin-1-stimulated calcium influx……………………………………………..121

5.3.5: DCC-ICD and transcription regulation…………………………………………122

5.4: Phosphorylation of DCC Thr 1210 regulates its stability………………………………...126

5.4.1: Phosphorylation as a protective mechanism……………………………………127

5.4.2: Phosphorylation modifies protein conformation………………………………..128

5.4.3: Identity of the kinase phosphorylating Thr 1210……………………………….129

5.4.4: The cytoplasmic tail of DCC is required for downstream signaling

and axon outgrowth……………………………………………………………130

5.5: Conclusion………………………………………………………………………………...130

Chapter 6: Materials and Methods…………………………………………………………...132

References…………………………………………………………………………………….140

viii

LIST OF FIGURES

Figure 1.1: Four primary areas in the rodent neocortex………………………………………….5

Figure 1.2: Cortical projection neuron subtypes…………………………………………………7

Figure 1.3: Cortical interneuron subtypes………………………………………………………..8

Figure 1.4: Development of the rodent neocortex………………………………………………12

Figure 1.5: Polarization of cortical neurons in vitro…………………………………………….15

Figure 1.6: Polarization of cortical neurons in vivo…………………………………………….23

Figure 1.7: The neuronal growth cone………………………………………………………….26

Figure 1.8: The road signs throughout the road trip…………………………………………….29

Figure 1.9: Commissural axon guidance in the spinal cord…………………………………….37

Figure 1.10: Structural domains of netrin………………………………………………………43

Figure 1.11: DCC is a type I transmembrane receptor part of the immunoglobulin

superfamily………………………………………………………………………..45

Figure 1.12: Netrin-1 associates with two DCC molecules……………………………….……46

Figure 1.13: DCC mutation landscape identified in the human population…….………………51

Figure 1.14: Cortical axon guidance…………………………………………………..………..58

Figure 1.15: The long journey along the cortical spinal track……………………………..…...59

Figure 1.16: Domain structure of the classical calpains………………………………………...62

Figure 2.1: The N-terminal SH2 domain of p120RasGAP interacts in vitro with DCC via phosphorylated tyrosine Y1418………………………………………………………………...68

Figure 2.2: p120RasGAP interacts with DCC in netrin-1-induced embryonic cortical neurons………………………………………………………………………………………….69

Figure 2.3: p120RasGAP associates with a DCC multiprotein complex in netrin-1-stimulated cortical neurons……………………………………………………………71 Figure 2.4: p120RasGAP is required to maintain basal Ras and ERK activities in cortical

neurons………………………………………………………………………………………….73 Figure 2.5: p120RasGAP is required for netrin-1-dependent attraction………………………..76 Figure 2.6: The N-terminus of p120RasGAP is sufficient to mediate netrin-1-dependent cortical axon outgrowth…………………………………………………………………………79 Figure 2.7: Human p120 RasGAP is expressed in p120RasGAP-deficient rat embryonic cortical neurons………………………………………………………………………………....80 Figure 2.8: The N-terminus of p120RasGAP interacts with the C-terminus of DCC………….83 Figure 3.1: Expression of calpain-1, calpain-2, calpastatin and the netrin receptor, DCC during cortical neuron development in vitro……………………………………………………85 Figure 3.2: Expression of calpain-1, calpain-2, calpastatin and the netrin receptor, DCC, have distinct but overlapping subcellular localization during cortical axon outgrowth at 2 days in vitro (DIV2)…………………………………………………………………………………….86

Figure 3.3: Netrin-1 stimulation increases calpain-1, but not calpain-2, expression while decreasing calpastatin levels……………………………………………………………………88

Figure 3.4: Netrin-1 activates calpain during cortical neurons axon outgrowth and axon specification………………………………………………………………………………….....91

Figure 3.5: Netrin-1-mediated calpain activation is ERK1/2-dependent………………………93

Figure 3.6: Netrin-1 activates calpain leading to proteolysis of its receptor DCC……………..95

Figure 3.7: Calpain limits cortical axon outgrowth…………………………………………….97

Figure 3.8: Calpastatin promotes axon outgrowth by inhibiting calpain1/2 activity…………...99

Figure 4.1: Identification of three novel Ser/Thr phosphorylation sites on the cytoplasmic tail of DCC………………………………………………………………………………………...101

Figure 4.2: The negative charge provided by the phosphorylation of T1210 is sufficient to protect DCC from proteolysis…………………………………………………………………102

Figure 4.3: DCC-T1210V and DCC-T1210E are targeted to the plasma membrane………...103

Figure 4.4: DCC ectodomain shedding in the media………………………………………….104

Figure 4.5: DCC full-length promotes cortical axon outgrowth………………………..……..105

Figure 4.6: Cleavage site of DCC-T1210V mutant……………………………………..……106

Figure 4.7: Calpastatin partially rescues cleavage of DCC-T1210V………………………….107

Figure 5.1. Proposed model for the role of p120RasGAP in netrin-1/DCC-mediated axon outgrowth and guidance………………………………………………………………………112 Figure 5.2: Proposed model for the role of calpain in netrin-1/DCC-mediated axon outgrowth and guidance……………………………………………………………………………...... ….114

Figure 5.3: Axonal projections in CAPNS1-Nestin-cKO brain……………………………….118

Figure 5.4: Netrin-1-mediated ERK1/2 activation is abolished in CAPNS1-Nestin-cKO cortical neurons…………………………………………………………………………..……120

Figure 5.5: Drosophila frazzled and rat DCC P3 domain alignment...... 123

Figure 5.5: Cleavage events downstream of netrin-1 stimulation…………….…………...….125

LIST OF ABBREVIATIONS

4EBP eIF-4E-binding protein 1 5HT3aR Ionotropic serotonin receptor A2b adenosine 2B receptor AC anterior commissure ACC agenesis of the corpus callosum ADAM A Disintegrin and Metalloprotease AIS axon initial segment BDNF brain-derived neurotrophic factor BMP bone morphogenetic factor C region central region Ca2+ calcium CAM cell adhesion molecule cAMP cyclic adenosine monophosphate CC corpus callosum CFuPN corticofugal projection neuron CGE caudal ganglionic eminence cGMP cyclic guanosine monophosphate CMM congenital mirror movement CNS central nervous system CPN callosal projection neuron CRIB Cdc42/Rac1 interactive binding Comm Commissureless CSMN Corticospinal motor neuron CST Corticospinal tract

xii

C-terminus carboxy terminus CThPN corticothalamic projection neuron CTPN corticotectal projection neuron DAG diacylglycerol DCC Deleted in Colorectal Cancer DLC1 deleted in liver cancer 1 DIV days of culture in vitro DSBS developmental split brain dSTORM direct stochastic optical reconstruction microscopy DSCAM Down Syndrome Cell Adhesion Molecule ECM extracellular matrix EGF epidermal growth factor Ena/VASP enabled/ vasodilator- stimulated phosphoprotein ER endoplasmic reticulum ERM ezrin-radixin-moesin ERK extracellular signal-regulated kinase ESC embryonic stem cell ExAC Exome Aggregation Consortium FAK focal adhesion kinase F-actin filamentous actin FGF fibroblast growth factor FN fibronectin GAP GTPase-activating protein GDNF Glial Cell Line-Derived Neurotrophic Factor GEF guanine nucleotide exchange

xiii

factor GFP green fluorescent protein GPI glycophosphatidylinositol GST glutathione S-transferase ICD intracellular domain Ig immunoglobulin IGF-1 insulin-like growth factor 1 IN interneuron IPC intermediate progenitor zone IZ intermediate zone LRR leucine-rich repeat MAP microtubule-associated proteins MGE medial ganglionic eminence MMP metalloprotease mRNA messenger RNA MT microtubule NGF nerve growth factor Nrp neuropilin NT-3 neurotrophin-3 N-terminus amino terminus NTR netrin-like domain N-WASP neural Wiskott-Aldrich syndrome protein P region peripheral region PAK p21 activated protein PALM photoactivated localization microscopy PDGF platelet-derived growth factor PI3K phosphatidylinositol 3-kinase pERM phosphorylated ERM proteins PIP2 phosphatidylinositol 4,5-

xiv

bisphosphate PIP3 phosphatidylinositol 3,4,5- triphosphate PKA cAMP-dependent protein kinase PKC protein kinase C PlTPα phosphatidylinositol transfer protein-α PN projection neuron PNS peripheral nervous system PTEN phosphatase and tensin homolog deleted on chromosome 10 PV parvalbumin RA retinoic acid RAP1B Ras-related protein 1B RBD Rho binding domain RG radial glia RPTP receptor protein tyrosine phosphatase RGC retinal ganglion cell Rho Ras homologous ROCK Rho kinase ROI region of interest R-Ras Related to Ras RTK Receptor tyrosine kinase ScPN subcerebral projection neuron SEM standard error of the mean Sema3A semaphorin 3A SFK Src family kinase SH2 Src homology 2 SH3 Src homology 3

Shh sonic hedgehog SIM Structured Illumination Microscopy Smo smoothened SNP small nucleotide polymorphism SNARE soluble N-ethylmaleimide- sensitive factor attachment protein receptor SNAP25 synaptosomal-associated protein of 25 kDa Sos son of sevenless SST somatostatin STED Stimulated Emission Depletion SVZ subventricular zone TAG-1 transient axonal glycoprotein-1 TGFβ transforming growth factor β TRIM Tripartite Motif Containing TRITC tetramethyl rhodamine isothiocyanate TUBB3 βIII-tubulin T zone transition zone UNC uncoordinated VAMP vesicle-associated membrane protein VEGF Vascular Endothelial Growth Factor VZ ventricular zone WASP Wiskott-Aldrich syndrome protein WAVE WASP-family verprolin- homologous protein ZBP1 Zipcode binding protein 1

xvi

ABSTRACT

Neural circuits are the basis of the human experience. The long journey from birth to becoming a functional unit within the brain neural circuitry encompasses several processes which are the research interest of distinct neuroscientific sub-disciplines; from 1) neural induction, to 2) neurogenesis, followed by 3) neuronal polarization, 4) axon outgrowth and guidance, to 5) synaptogenesis, and finally, 6) pruning of excess connections. Following neurogenesis, newly- born neurons polarize and initiate axon outgrowth toward their future synaptic partner.

Specialized embryonic structures, serving as intermediate targets throughout their journey, secrete guidance molecules that act as traffic lights, while adhesive molecules pave the road ahead. The molecular mechanisms mediating the cell’s behavior in response to these signaling molecules is the subject of this thesis. Specifically, this thesis examines the molecular mechanisms involved in regulating the receptor for netrin-1, deleted in colorectal cancer (DCC), during cortical axon outgrowth and guidance.

DCC becomes phosphorylated in response to netrin-1 binding. The phosphorylation of the conserved tyrosine 1418 (Y1418) is required for netrin-1-mediated axon outgrowth and guidance in vertebrates. Here we identified and characterized a novel SH2-containing binding partner to the phosphorylated Y1418, p120RasGAP. We show that upon netrin-1 stimulation, p120RasGAP is recruited to the DCC signaling platform and is required for netrin-1-mediated axon outgrowth and guidance.

In addition, netrin-1 binding to DCC induces receptor proteolysis. Here we identify a novel protease, calpain, activated in response to netrin-1 which cleaves DCC. We show that activation of calpain downstream of netrin/DCC is ERK1/2-dependent and essential for normal axon outgrowth.

xvii

Finally, since the majority of DCC phosphorylation occurs on ser/thr residues, we wanted to identify and investigate their function. One of these sites, threonine 1210 (T1210), which was found to be mutated in humans, regulates DCC stability. The negative charge provided by the phosphorylation modification prevents cleavage and maintains DCC in its full-length form.

Removing this negative charge renders DCC susceptible to proteolysis and unable to induce cortical axon outgrowth. Taken together, the novel findings presented in this thesis contribute to and improve our knowledge of the mechanisms regulating DCC and the netrin/DCC signaling pathway during cortical development. The human diseases involving DCC reach well beyond the nervous system and preclinical trials are underway targeting the interaction between netrin-1 and DCC. Therefore, increasing our understanding of the mechanisms regulating DCC in health will undoubtedly improve our therapeutic strategies in disease.

xviii

RÉSUMÉ

Les circuits neuronaux forment la base de l’expérience humaine. Le long périple de la naissance jusqu’à devenir une unité fonctionnelle de la circuiterie neuronale du cerveau comprend plusieurs processus faisant l’objet de sous-discipline distincte dans le domaine des neurosciences : de 1) l’induction neuronale à 2) la neurogénèse, suivi de 3) la polarisation neuronale, 4) la croissance et le guidage axonal, jusqu’à 5) la synaptogénèse, et finalement 6) l’élimination du ‘surplus’ de connections. Suivant la neurogénèse, les neurones nouvellement nés polarisent et initient la croissance axonale vers leurs futurs partenaires synaptiques. Des structures embryonnaires spécialisées jouant le rôle de cibles intermédiaires durant leurs voyages sécrètent des molécules de guidages agissants comme des feux de circulation alors que des molécules d’adhésions pavent la route devant. Les mécanismes moléculaires à la base des comportements cellulaires en réponse à ces molécules de guidage font l’objet de cette thèse. En particulier, cette thèse examine les mécanismes moléculaires impliqués dans la régulation du récepteur pour la molécule de guidage axonal, nétrine-1, deleted in colorectal cancer (DCC) lors de la croissance et le guidage axonal des neurones corticaux.

DCC devient phosphorylé suivant sa liaison avec nétrine-1. La phosphorylation de la tyrosine conservée 1418 (Y1418) est requise pour la croissance et le guidage axonale induit par la nétrine-1 chez les vertébrés. Nous avons identifié et caractérisé un nouveau partenaire de liaison se liant avec la tyrosine phosphorylée de DCC par son domaine SH2, p120RasGAP.

Nous démontrons que suivant la stimulation avec la nétrine-1, p120RasGAP est recruté vers la plate-forme de signalisation DCC et est requis pour la croissance et le guidage axonal dirigé par nétrine-1.

xix

De plus, la liaison de la nétrine-1 à DCC induit la protéolyse du récepteur. Nous identifions une nouvelle protéase activée en réponse à la nétrine-1, calpain, qui clive DCC. Nous démontrons également que l’activation de calpain en aval de nétrine-1/DCC est dépendant d’ERK1/2 et est essentiel à une croissance axonale optimale.

Finalement, puisque la majorité de la phosphorylation de DCC se fait sur les résidus

Ser/Thr, nous identifions quatre nouveaux sites de phosphorylations. Un de ces sites, Thr 1210

(T1210), qui est également muté chez la population humaine, régule la stabilité de DCC. La charge négative procurée par la phosphorylation prévient le clivage et maintient DCC dans sa forme complète. Une fois cette charge négative éliminée, DCC est plus susceptible à la protéolyse et incapable d’induire la croissance axonale. Pris ensemble, ces nouvelles découvertes présentées dans cette thèse contribuent et améliorent nos connaissances des mécanismes régulant DCC et la voie de signalisation nétrine-1/DCC lors du développement cortical. Les maladies humaines impliquant DCC s’élargissent plus loin que le système nerveux et des essais précliniques sont en cours ciblant l’interaction entre la nétrine-1 et DCC. Donc en améliorant notre compréhension des mécanismes régulant DCC chez l’individu en santé nous allons sans doute pouvoir améliorer nos stratégies thérapeutiques dans le contexte de la maladie.

This is a traditional thesis with the results section split into 3 chapters

ORIGINAL SCHOLARSHIP AND CONTRIBUTIONS TO

KNOWLEDGE

Original contributions (presented as Chapters 2, 3, and 4, respectively): 1. “p120RasGAP Contributes as a Scaffolding Protein to Regulate Netrin-1-Mediated Axon Outgrowth and Attraction in a GAP-dependent Manner”, Antoine-Bertrand, J, Duquette P.M., Alchini R., Li X., Kennedy T.E., Fournier A.E., and Lamarche-Vane, N. The journal of biological chemistry 2016 Feb 26;291(9):4589-602.

This study is the first report on the regulation and function of p120RasGAP during netrin- 1/DCC signaling and demonstrates that p120RasGAP is required for netrin-1/DCC- dependent axon outgrowth and chemoattraction. It also provides the first direct evidence that Ras is activated downstream of netrin-1.

2. “A role for the calcium-activated protease calpain in the regulation of netrin-1/DCC- mediated cortical axon outgrowth and guidance”, Duquette PM & Lamarche-Vane N, Manuscript ready for submission

This study provides the first description of calpain as protease cleaving and regulating the netrin-1/DCC signaling pathway during cortical axon outgrowth. This study is also the first to identify, in mammalian neurons, the intracellular domain (ICD) of DCC which increases in response to netrin-1 stimulation. Future studies will help determine the function of this DCC-ICD.

3. “Investigating the role of DCC-Thr 1210 phosphorylation during cortical axon outgrowth”, Duquette PM, Piché C., Luangrath V., and Lamarche-Vane N.

xxi

This study offers the first report of a phosphorylation site necessary to maintain DCC in its full length form. It also shows that phosphorylation of Thr 1210 is required for downstream signaling and to promote cortical axon outgrowth. Preliminary results indicate that this phosphorylation might be protecting DCC from calpain-mediated proteolysis. Published manuscripts not presented in this thesis

4. “Rho GTPases in embryonic development”, Duquette PM & Lamarche-Vane N. Small GTPases. (2014);5(2):8

(peer-reviewed review article)

5. “CUX2 protein functions as an accessory factor in the repair of oxidative DNA damage” Pal R, Ramdzan ZM, Kaur S, Duquette PM, Marcotte R, Leduy L, Davoudi S, Lamarche-Vane N, Iulianella A, and Nepveu A. The journal of biological chemistry (2015); 290(37):22520-31.

6. “CdGAP/ARHGAP31, a Cdc42/Rac1 GTPase regulator, is critical for vascular development and VEGF-mediated angiogenesis”, Caron C, DeGeer J, Fournier P, Duquette PM, Luangrath V, Ishii H, Karimzadeh F, Lamarche-Vane N, Royal I. Scientific reports (2016);6:27485

xxii

CONTRIBUTIONS OF AUTHORS

CHAPTER 2: p120RasGAP Contributes as a Scaffolding Protein to Regulate Netrin-1- Mediated Axon Outgrowth and Attraction in a GAP-dependent Manner

Judith Antoine-Bertrand: Designed, executed and analyzed all experiments. Assembled all the figures and co-wrote the manuscript. Philippe M. Duquette: Contributed to the execution of experiments for Figures 2.4 and 2.3. Ricardo Alchini: Contributed to the design, execution and analysis of the Dunn chamber assay in Figure 2.5. Timothy E. Kennedy: Provided the recombinant netrin-1 VI-V in Figure 2.5. Alyson E. Fournier: Contributed to the design and analysis of the Dunn chamber assay in Figure 2.5. Nathalie Lamarche-Vane: Designed and analyzed all experiments. Co-wrote the manuscript.

CHAPTER 3: A role for the calcium-activated protease calpain in the regulation of netrin-1/DCC-mediated cortical axon outgrowth and guidance

Philippe M. Duquette: Designed, executed and analyzed all experiments. Assembled all the figures and co-wrote the manuscript. Nathalie Lamarche-Vane: Designed and analyzed all experiments. Co-wrote the manuscript.

CHAPTER 4: Investigating the role of DCC-Thr 1210 phosphorylation during cortical axon

Outgrowth

Philippe M. Duquette: Designed, executed and analyzed all experiments except for Figure 4.1. Assembled all the figures. Vilayphone Luangrath: executed and analyzed experiments in Figure 4.1 Chantal Piché: Designed and executed part of the experiments in Figure 4.1

xxiii

Nathalie Lamarche-Vane: Designed and analyzed all experiments.

xxiv

Preface to Chapter 1

This chapter is a literature review of the mechanisms regulating neuronal polarization and axon specification, outgrowth, and guidance during development of the central nervous system with an emphasis on cortical development. The functions of the growth cone cytoskeleton downstream of classical guidance cues are presented. Finally, the strategies involved in the regulation of guidance receptors are presented with an introduction of a novel protease implicated in netrin/DCC signaling, calpain.

Chapter 1: Introduction and literature review

“Men ought to know that from the brain, and from the brain only, arise our pleasures, joys, laughter and jests, as well as our sorrows, pains, griefs and tears. Through it, in particular, we think, see, hear, and distinguish the ugly from the beautiful, the bad from the good, the pleasant from the unpleasant. . . .” Hippocrates, 5th Century, B.C.1

The brain is arguably the most complex structure to have ever evolved. How does the brain activity at the cellular and molecular level create what Plato defined as the Good, the True, and the Beautiful over 2500 years ago?2 These three characteristics which define the human experience and what we stride towards in our everyday lives can be reduced to, according to modern neuroscience, the sum of the activity of each neuron within our neural network. Each thought and experience is the result of a specific combination of neural activity that creates the world we are ‘immersed’ in. This world is not independent of us and each creates its own…3-6

To understand the process by which the right kinds of neurons are produced and connected to each other is not only fundamental to understanding how our brain activity at the neural network level mediates our thoughts and behaviors but more importantly, how our world is created.

1.1 - The Nervous System

“As long as our brain is a mystery, the universe, the reflection of the structure of the brain will also be a mystery.” Cajal7

The nervous system is responsible for relaying information from both the external environment and the internal state of homeostasis within multicellular organisms. This information is then assimilated to coordinate a response (behavior).8, 9 Our current understanding of the nervous system and its structural elements stems from two fundamental discoveries made by Camilo Golgi’s silver stain and Santiago Ramón y Cajal’s histology. The first picture of neurons was made by Golgi’s pioneering method for staining neurons. Using

Golgi’s method, Cajal investigated the histology of the nervous system that led him to conclude

2 that the brain and spinal cord were composed of discrete elements, neurons, that may touch one another but do not fuse. This came to be known as the ‘neuron doctrine.’ The term ‘neuron’ was coined by Heinrich Wilhelm Gottfried von Waldeyer-Hartz to refer to the nerve cell. Earlier theories of brain organization postulated that the structural elements were fused in a net-like meshwork; this was before the electron microscope and the discovery of the plasma membrane that separates each cell.10 Eventually the space (or “zones for the transmission of stimuli”) between neurons was termed ‘synapse’ by Sir Charles Scott Sherrington and therefore supported

Cajal’s theory.11 Cajal’s work,7 which is still regularly consulted by modern-day scientists remains one of the greatest contributions, not only to neuroscience, but to science in general.

1.1.1 - The Cell Types of the Nervous System 1.1.1.1 - Neurons Neurons (nerve cells) are among the most polarized cell type in the body.12, 13 With its dendritic tree receiving input signals from neighboring cells through a specialized junction called a synapse, it sends a signal (action potential) of its own through its axon, depending on the strength of the many signals it receives from its synaptic partners.9, 14 Integration of all input signals is done at the level of the cell body where the decision to propagate or terminate the signal is made. Once the threshold has been reached to initiate an action potential down its axon, an electrical signal mediated by an influx of positive sodium and efflux of positive potassium ions and negative chloride ions is sent along its axon terminal where the electric signal is converted into a chemical one.8, 9 Axon terminals are specialized regions at the tip of axons where specialized vesicles are ready to release their content into the synapse to bind specific receptors located on the postsynaptic side of the synapse.8, 9, 14 These specialized regions of the axon make up the presynaptic side of the synapse and are enriched in several proteins involved in

3 synaptic transmission.8, 9, 14 Conversely, during development, the axonal growth cone located at the tip of the extending axon is also a specialized structure serving a different function. The highly motile, fan-shaped, axonal tip explores and guides the axon along its journey to its synaptic partner.15

1.1.1.2 - Glial cells Although glia cells have traditionally been labelled as supportive cells for the main stars of the nervous system, neurons; they make up 90% of the cellular population in the human brain.16, 17 Some of their functions are to insulate axons to increase the speed and efficiency of the electrical signal by forming a myelin sheath around axons similar to the way electric wires are insulated by rubber.8, 9 Oligodendrocytes are responsible for insulating axons of the CNS while Schwann cells accomplish this in the PNS.8, 9, 18 The third glial cell-type are astrocytes.

They serve several homeostatic functions such as the establishment of the blood-brain barrier and regulation of ion concentration.8, 9, 18, 19 Astrocytes are also involved in signaling to neurons by secreting various growth factors and regulating calcium signaling between neurons. Radial glial cells located in the ventricular zone of the developing telencephalon (see below) also serve important functions during neurogenesis and neuron migration, which I will come back to below.20-23 Oligodentrocytes, Schwann cells, and astrocytes together make up the macroglia while microglia are immune cells within the nervous system.18

1.1.2 - The Central Nervous System The CNS consists of the spinal cord and the brain. Starting from the base of the skull, the spinal cord is the most caudal (posterior) part of the CNS and the most accessible. As we will see below the spinal cord has been widely used as a model to study axon guidance due to its

4 relative simplicity and accessibility for manipulations.24 It receives sensory information that is relayed to the brain for integration where the response is initiated and sent back down the spinal cord to create a motor response.8, 9 Alternatively, reflex movements bypass the brain to allow a faster response time when minimizing the response time is crucial for survival (ex: removing your hand from a burning stove). The brain can be divided into six major regions: 1) medulla,

2) pons, 3) midbrain, and 4) diencephalon, which together make up the brain stem.8, 9 The brain stem is responsible for controlling basic functions such as blood pressure, respiration, and sleep.

Evolutionary speaking, it is the oldest part of the brain and is thus commonly called the reptilian brain.25 The fifth subdivision, the cerebellum (“small brain” in latin) located in the back of the head, contains more neurons than any other part of the brain and is responsible for the control and coordination of motor movements as well as being involved in various cognitive tasks.26

Finally, the most evolutionary recent addition to the CNS is the telencephalon, which includes the two cerebral hemispheres. The cerebral cortex, located dorsally, is responsible for higher cognitive functions and is the central governor of behavior, where information is integrated.8, 9, 23,

1.2 - Cellular Diversity in the Cortex 1.2.1 - Arealization It is in the cerebral cortex that we see the most diversity in neuronal subtypes.18, 21, 28, 29

Although the cerebellum contains the greater number of neurons, its neuronal population is relatively homogenous in comparison. The divisions of the neocortex into distinct functional areas30-32 were first introduced by the neuroanatomist Paul Broca, which we owe the Broca Area

(or Broca’s language area).8, 9 The rodent neocortex can be divided into four distinct primary areas: 1) somatosensory, 2) auditory, 3) visual, and 4) motor (Figure 1.1).21, 29 Each of the first

three primary areas is dedicated in processing information coming from the PNS while the motor

cortex, as the name implies, is responsible for initiating the command for movements that will go

down the spinal cord and to the motor neurons of the PNS, to stimulate the appropriate muscle.

Outside of these well-defined primary areas, other regions are mostly labelled “associative” for

lack of any yet identified area-specific functions. These areas tend to be activated under various

different tasks and relay information between the well-defined areas.

olfactory bulb neocortex Cerebellum anterior

Spinal Cord posterior

Figure 1.1: Four primary areas in the rodent neocortex

Schematic representation depicting the left hemisphere of the rodent neocortex. The olfactory bulb is the most anterior structure (upper left) with the spinal cord being the most posterior structure (bottom right). The four primary areas: 1) motor cortex (blue), 2) somatosensory (green), auditory (yellow), and visual (red). Adapted with permission from Annu. Rev. Cell Dev. Biol. 2015 (Lodato & Arlotta, 2015). Copyright © 2015 by Annual Reviews.

1.2.2 - Nomenclature Within the various areas of the cortex, the cell bodies of neurons are organized into 6

layers where each layer shares common connectivity with other areas of the CNS.18, 20, 21, 29, 33

The layers are organized such that as you move from the skull deeper into the brain you

encounter different layers of neurons. The superficial layers (I/II) are closest to the skull while

the deep layers are the furthest from the skull and closest to the inner ventricles. The great

endeavour of classifying all cortical neurons started over a hundred years ago and remains

incomplete. Employing the method developed by Camillo Golgi, Santiago Ramon y Cajal –

6 widely recognized as the Father of Neuroscience, was able to sketch some detailed drawings of the cerebral cortex, highlighting its great heterogeneity and cellular diversity.7, 34-36 Today’s nomenclature of cortical neurons is still largely based on connectivity which was first proposed by Cajal in 1909,34 although molecular characterization through gene expression profile, is now adding some complexity in the identification of the various cortical neuron subtypes. Cortical neurons are now defined by a combination of characteristics such as molecular markers, morphology, and electrophysiological characteristics, in addition to their connectivity.18, 21, 29, 37

1.2.3 - Cortical neuron subtypes – projection neurons There are two types of neurons in the cerebral cortex: projection neurons (PN) and interneurons (IN) organized in six distinct layers. Projection neurons are excitatory glutamatergic and connect the cerebral cortex to the rest of the CNS. They make up around 80% of the neuronal population in the cortex and make long range connections.18, 21, 29 Although shorter than axons of the sciatic nerve that can measure more than one meter, PN typically have long axons that can measure tens of centimeters. They can be classified as either intracortical or corticofugal.18, 21, 29, 37, 38 Intracortical neurons are predominantly located within superficial layers II/III. They are either “associative” and establish connections within the same hemisphere or “commissural” and cross the midline to establish connections with neuron in the other hemisphere. The axons of these commissural projection neurons project to the contralateral side through two commissures: the anterior commissure (AC) and the corpus callosum (CC).18, 21, 29

The AC is the most evolutionary conserved commissure but is responsible for a small minority of crossing axons in rodents and humans. Most commissural neurons cross the midline through the corpus callosum and are thus termed callosal projection neurons (CPN) (Figure 1.2A &

1.4).18, 21, 39 Corticofugal PNs (CFuPNs) are located in deep cortical layers (V-VI) and send their

axons to distal targets outside the cortex. They can be subdivided into corticothalamic (CThPNs)

in layer VI (Figure 2B) and subcerebral (ScPNs) in layer V (Figure 1.2C & 1.4). ScPNs project

mainly to the pons and other nuclei of the brainstem (corticopontine), to the superior colliculus

(corticotectal), and the spinal cord (corticospinal motor neurons) (Figure 1.2C & 1.4). Together

with this basic framework, each of these cortical neuron subtypes has been further characterized

at the molecular level over the past 15 years. Thus, the nomenclature based on connectivity

initially proposed by Cajal together with differences in morphology and electrophysiological

properties is now being combined with the specific expression of distinct molecular markers.

Thus, the molecular profiling of cortical neurons enabled the identification of novel class-

specific signature genes and identified distinct cortical neuron subtypes (Figure 1.2).

A B C

Figure 1.2: Cortical projection neuron subtypes

(A) Callosal projection neurons (CPN) located in superficial layers are commissural neurons of the

neocortex establishing connections with neurons on the contralateral hemisphere through the corpus

callosum (CC). A list of molecular markers characterizing this subtype is shown on the right. (B) Figure 2: Corticothalamic projection neurons (CThPN) are a subset of corticofugal neurons projecting to the thalamus and mainly located in layer VI. Some molecular markers identified are shown on the right (C) Subcerebral projection neurons (ScPNs) located in layer V send their axons outside the cerebral

hemispheres to the superior colliculus (coritcotectal – CTPN), to the brain stem (corticopontine) and

spinal cord (corticospinal motor neurons – CSMN). Some molecular markers shared by these neuronal subtypes are shown on the right. Adapted with permission from Annu. Rev. Cell Dev. Biol. 2015 (Lodato & Arlotta, 2015). Copyright © 2015 by Annual Reviews.

1.2.4. - Cortical neuron subtypes – interneurons With only local connections, interneurons have not been characterized based on connectivity and no universally accepted classification exists. Some classification attempts have been made based on molecular markers, electrophysiological properties and dendritic tree morphologies.18, 40 Briefly, three main subtypes have been identified based on non-overlapping expression of parvalbumin (PV+ - 40% of INs), somatostatin (SST+ - 30% of INs), and the ionotropic serotonin receptor (5HT3aR+ - 30% of INs)(Figure 1.3 & 1.4). Morphologically,

PV+ are subdivided into basket and chandelier cells while SST+ INs are mainly Martinotti cells

(Figure 1.3) . INs are heterogenous and further subdivisions based on combinations of other markers and electrophysiological/morphological differences have been used which are beyond the scope of this introduction.41, 42

Figure 1.3: Cortical interneuron subtypes

Three main subtypes of interneurons based on non-overlapping expression of parvalbumin (PV), somatostatin (SST), and the ionotropic serotonin receptor (5HT3aR). Within these broad subtypes, they can be further categorized based on morphology (basket, chandelier, uni-, bi-, or multipolar). Adapted with permission from Annu. Rev. Cell Dev. Biol. 2015 (Lodato & Arlotta, 2015). Copyright © 2015 by Annual Reviews.

1.3 - From Neural Induction to Cortical Assembly 1.3.1 - Neural Induction “Vertebrate embryonic cells will become nerve cells unless told otherwise” is the title of a 1997 Cell review that summarizes the default model of neural induction.43, 44 It was the pioneer transplant experiments of Spemann and Mangold that led to the identification of the

“organizer”, an embryonic structure responsible for inducing neural tissue.44-46 By grafting a piece of dorsal mesoderm taken from an early newt gastrula on the ventral side of a second embryo, Spemann and Mangold noticed the appearance of a well-organized second nervous system developing on the ventral side of the host embryo. The dorsal mesodermal structure was later termed the Spemann organizer for its neural induction properties. In rodents, the equivalent structure is the node. Following gastrulation, the embryo consists of three germ layers: endoderm, mesoderm, and ectoderm.44, 47 Interestingly, it is the absence of instructive signals in the dorsal ectoderm that enables neural induction; the entry of multipotent ectodermal cells into the neural lineage. Among the molecules secreted from the node and responsible for neural induction in vivo are all inhibitors of BMP signaling: noggin, chordin, follistatin, and cerberus.

Thus, it is the inhibition of BMP signaling in the dorsal ectoderm by the underlying mesoderm that is permissive to the development of the nervous system. Furthermore, in vivo ectodermal cells or embryonic stem cells (ESCs) cultured in vitro acquire a forebrain-specific neural fate in the absence of any factors added to the culture media. After neural induction, the dorsal ectoderm (i.e. neuroectoderm – neural plate) consists of a layer of neuroepithelial cells that will roll into a tube to form the neural tube, a process called “neurulation”. The anterior part of this tube will become the forebrain in the absence of any signal while the acquisition of a more posterior identity (i.e. midbrain, hindbrain, and spinal cord), need additional factors such as Wnt,

FGF8, and Retinoic acid (RA).44 Being the dorsal part of the forebrain, the cerebral cortex is one

10 of a few structures that emerge from ESCs in the absence of any inductive signals. However, additional signals are required within the forebrain to fine-tune the differentiation into cortical neurons rather than other ventral structures of the forebrain such as the striatum. Throughout the anterior-posterior axis of the CNS, sonic hedgehog (Shh) is secreted ventrally to form a gradient along the dorso-ventral (DV) axis.48 Depending on the location along this axis, progenitor cells will be exposed to different concentrations of Shh and will respond by activating the expression of different transcription factors that will lead to the expression of subtype-specific genes. Thus,

Shh secreted from the ventral plate acts as a morphogen that competes with other morphogens secreted from dorsal structures (i.e. BMPs, Wnts).48, 49 Inhibition of Shh signaling is therefore required to induce cells into the cortical lineage (i.e. dorsal identity). Taken together, the default model of neural induction tells us that ESCs are destined to become a forebrain but don’t due to signals instructing them to become other structures. Therefore to become a cortical neuron, no

“instructions” are necessary; it’s the ESC’s destiny.

1.3.2 - Cortical Assembly Once the neural tube has closed, neurepithelial cells begin to differentiate into radial glia

(RG) that will establish the ventrical zone (VZ). Radial glia acts as stem cells and can either divide asymetrically into neurons or intermediate progenitor cells (IPCs) that will migrate and establish a secondary proliferative region, the subventricular zone (SVZ) (Figure 1.4).18, 20, 22, 23,

28 Each RG division self-renews to maintain the RG population while most IPCs divide into two postmitotic neurons that will migrate along RG into the cortical plate. As mentioned above, RG are involved in the migration of newly-born neurons. By spanning the entire thickness of the cortex, RG are used as scaffold by newly-born neurons to migrate into the cortical plate, their final destination where they will terminally differentiate. Neurons migrate in an inside-out

11 fashion such that early-born neurons populate deeper cortical layers while late-born neurons migrate past them to populate progressively superficial layers (Figure 1.4). From an evolutionary perspective, the SVZ, and its expansion in higher primates enabled the significant increase in the surface area (gyrification) and thickness of the cortex that expanded the relative number and diversity of cortical neurons.50, 51 This led to the acquisition of higher cognitive functions. At the cellular level, human cortical neurons display distinct patterns of neuronal morphology due to an expanded SVZ leading to a longer relative proliferative period and neurogenic window. This is accompanied with a more prolonged period of neurite outgrowth, dendritic spine maturation, and synaptogenesis.21, 52 Taken together, the prolonged developmental period at the cellular level correlates well with the increase weaning period and dependency of newly-born human babies on their parents. Interneurons, on the other hand, are born outside the developing cortex and have to migrate long distances tangentially from the medial and caudal ganglionic eminence (Figure 1.4).53 At the molecular level several transcription factors have been identified as necessary for the generation of the cortical neuron subtypes mentioned above (Figure 1.2).18, 20, 21, 23, 29, 38, 52 Briefly, Fezf2 is required for the generation of SCPN (subcerebral projection neurons) as no cortical neurons project to the brain stem and spinal cord in Fezf2-null mice.54, 55 In these mice, Tbr1-positive CThPN expands into layer V, indicating a change in cortical neuron identity. Thus, Tbr1 acts in opposition to Fezf2, and its downstream target gene Ctip2, to specify CThPN by directly repressing Fezf2 expression.55 Conversely, SATB2 is necessary for specification of CPN and repression of

CFuPN.56-59 In Satb2-deficient mice, almost no axons cross the corpus callosum even though the midline appears normal while CPN-specific genes are either completely lost or severely reduced.56, 57, 59 Interestingly, it was shown that both receptors for the guidance cue netrin-1,

Unc5 and in particular deleted in colorectal cancer (DCC) - the focus of this thesis, are under the direct control of Satb2.60 This highlights the importance of initiating the proper transcriptional program early to allow the expression of all necessary receptors necessary for normal axon outgrowth and guidance later in development. Finally, it is worth mentioning that CThPN and

SCPN neurons share a common lineage, CFuPN, and its upstream master regulator, Sox5.61 As such, their axons travel together for several days before their trajectory diverge and the expression of Tbr1 and CTIP2 becomes restricted to CThPN and SCPN, respectively. Taken together, signaling centers throughout the embryo secrete different factors (morphogens) that will establish different gradients.49 Cells exposed to different concentrations of theses morphogens will respond by activating different combination of transcription factors that will activate subtype-specific genes necessary to differentiate into distinct cortical neural subtypes.

Following neurogenesis, three key events must take place: 1) polarization through axon specification, followed by 2) axon outgrowth and guidance, and 3) branching and synaptogenesis

Figure 1.4: Development of the rodent neocortex Neurogenesis in mice (E11-E19) and rats (E13-E21) occurs entirely before birth. Radial glial cells (RG) spanning the entire thickness of the cortex serve as scaffhold for newly-born neurons. Later

in development, RGs generate intermediate progenitor cells (IPCs) giving rise to a secondary proliferative region, the subventricular zone (SVZ). At this stage, the intermediate zone (IZ) starts to enlarge, making room for axonal projections that have to pass through the IZ before exiting the cortex. Newly-born neurons migrate radially along RGs past earlier-born neurons to populate progressively superficial layers such that neurons located in superficial layers are born last.

Interneurons, on the other hand, are born outside the cortex within the ganglionic eminences (GE, medial – MGE and caudal – CGE) and have to migrate tangentially. Adapted with permission from Annu. Rev. Cell Dev. Biol. 2015 (Lodato & Arlotta, 2015). Copyright © 2015 by Annual Reviews.

1.4 - Neuronal polarization, axon specification, outgrowth, and guidance 1.4.1 - Neuronal Polarization The process of neuronal polarization has received so much attention over the last 30 years and tremendous progress has been made in characterizing the molecular mechanisms involved.62

Axons are characterized by a uniform distribution of microtubules with the plus-end facing away from the cell body while dendrites have a mix orientation.12, 63, 64 This orientation enables the polarized transport of various cargos involved in axon specification, differentiation, and maintenance.65 This enables dynein, a minus-end-oriented microtubule motor protein, to bring cargos from the axon back to the cell body while transporting other cargos from the cell body into dendrites. In addition, axonal microtubules are mainly decorated by Tau and MAP1B whereas dendritic microtubules are labelled by MAP2a-c.12, 63, 66 One of the pioneer papers that first proposed and characterized five distinct stages of neuronal polarization in vitro used cultured hippocampal neurons because of their homogeneity.62, 67 At stage 1, freshly plated immature neurons display highly dynamic actin cytoskeletal structures such as lamellipodia and

14 protrusive filopodia (Figure 1.5). Stage 2 is characterized by the emergence of multiple immature neurites occurring within the first 2 days in culture (Figure 1.5). Accumulation of axon determinants (Ras, cAMP/PKA, PI3K/PIP3, and pLKB1)68 in one neurite leads to a break in symmetry that triggers the first steps in axon specification at stage 3 (Figure 1.5). At this stage, one neurite starts to grow rapidly to become the axon (5-10 times faster than other neurites destined to become dendrites) in response to polarizing factors (ex: BDNF, laminin, TGF-β,

Wnt5a, Sema3A, and Netrin-1). Actin waves, which migrate from the cell body into one neurite, transiently widen the neurite shaft to allow increased microtubule polymerization, lead to bursts of neurite extension into the future axon.69 Orientation of organelles such as the Golgi, centrosomes, mitochondria, and endosomes correlates with the neurite that will become the axon.12 This normally starts at the beginning of the second day in vitro (DIV2) for hippocampal/cortical neurons. Axon specification depends on actin destabilization and microtubule stabilization within the nascent axon, thus enabling rapid growth.12, 63, 64, 66, 70 Stage

4 is characterized by rapid dendritic and axonal outgrowth (DIV4-15) (Figure 1.5) leading to terminally differentiated dendritic spines and the formation of the axon initial segment (AIS) at stage 5 (DIV15-25) (Figure 1.5). The AIS is critical for clustering voltage-dependent sodium channels that will initiate action potentials.71, 72 This specialized structure is characterized by an actin meshwork composed of spectrins and ankyrin B and serving as a diffusion barrier preventing dendritic proteins from entering the axon.71, 72 It is important to highlight the fact that these neurons which were dissected at embryonic day 14.5 were already polarized within the embryo and are thus repolarizing in vitro. It is thus possible that these neurons retain some of the molecular characteristics of polarized neurons and are not necessarily polarizing from scratch.12 I will focus on molecular events occurring at stages 2-3 (neurite outgrowth, axon

15 specification and outgrowth) since the focus of the lab is to characterize the molecular mechanisms involved in the netrin-1/DCC signaling pathway during axon outgrowth at DIV2.

Figure 1.5: Polarization of cortical neurons in vitro

First described by Dotti et al., (1988) in hippocampal neurons, cortical neurons follow a specific

sequence of morphological changes that characterizes polarization in dissociated cultures.

Lamellipodia and protrusive filopodia characterizes stage 1, leading to the formation of multiple immature neurites at stage 2. The critical step when symmetry is broken and one neurite starts to

grow rapidly to become the axon (axon specification- purple) happens during stage 3. Other

neurites will become dendrites (green). Rapid outgrowth of both the axon and dendrites

characterizes stage 4. Finally, terminal differentiation, which includes maturation of dendritic

spines and the formation of the axon initial segment (yellow), defines stage 5. Adapted with

1.4.2 - Mechansims regulating neuronal polarization 1.4.2.1 – Super-resolution and TRIM46 Although the bulk of our understanding of how neurons establish their polarity have been discovered in vitro over the last 30 years, much of the in vivo studies of the last decade have confirmed these findings and provided support for the established molecular mechanisms.73, 74

Intrinsic mechanisms exist that act in the absence of any external cues but the tissue environment play a dominant role in determining axon/dendrite polarity in vivo and will be discussed below.

Furthermore, new microscopy advances such as the different super-resolution techniques (i.e.

Structured Illumination Microscopy (SIM), Stimulated Emission Depletion (STED), and photo activated localization microscopy (PALM)/ direct stochastic optical reconstruction microscopy

(dSTORM) have allowed an even finer detailed description of the cytoskeletal structures involved at sub-200nm (limit imposed by light diffraction) resolution.75 By combining live-cell imaging, electron microscopy, and dSTORM super-resolution, one study identified the microtubule-associated protein TRIM46 as the master regulator of microtubule orientation by forming cross-bridges between parallel microtubules in the nascent axon.76 They show that

TRIM46 is required for axon specification and neuronal polarity in vitro and in vivo.76 They propose a model in which formation of uniform microtubules in the axonal shaft initiates neuronal polarity. Sub-diffraction microscopy techniques were also able to elucidate a novel actin structure within axons that went unnoticed by traditional light microscopy. In particular, circumferential rings of actin that wrap underneath the plasma membrane spaced at approximately 190 nm and connected by spectrin tetramers was revealed using STORM super- resolution microscopy.77 Thus, a unique feature of the submembranous axonal cytoskeleton is its periodic, ladder-like structure. This periodic organization of actin was also seen in live neurons using STED and are present early in the most proximal axon segment and propagate as the axon grows.77 Thus, similar to these examples, new super-resolution techniques will be able to characterize in finer details and identify novel structures within developing neurons to give us a better understanding of its molecular architecture.

1.4.2.2 -PI3K/PIP3 The activity of the lipid kinase phosphatidylinositol 3-kinase (PI3K) along with the accumulation of phosphatidylinositol (3,4,5)-triphosphate (PIP3) in the membrane of one neurite

17 seems to be necessary and sufficient for axon specification since blocking its activity prevents axon formation while expressing a constitutively active PI3K is enough to form multiple axons.12, 68, 78-80 It was shown that local application of laminin leads to local accumulation of

PIP3, suggesting that it lies upstream of PI3K, together with EGFR/Ras.78 Finally, two PI3K- interacting proteins, Shootin1 and Singar1/2 were also shown to act upstream of PI3K during axon specification.81, 82 In direct opposition to PI3K is the lipid and protein phosphatase PTEN

(phosphatase and tensin homolog deleted on chromosome 10) that dephosphorylates PIP3 into

PIP2.79, 80 RNAi-mediated PTEN knockdown leads to the formation of multiple axons, similar to the expression of constitutively active PI3K.79 Among the various proteins recruited to PIP3- enriched membrane regions through their PH domains, AKT appears to be crucial in axon specification.80, 83 Following translocation to the membrane, AKT is activated by PKD1/2 and

ILK-mediated phosphorylation. Expression of a constitutively active AKT is sufficient to induce multiple axons independent of PI3K, consistent with AKT acting downstream of PI3K.83

GSK3β, a negative regulator of axon formation and major hub in the polarity pathways, undergoes inactivating phosphorylation by AKT/ILK.84-86 This inactivation in turn relieves inhibition of several microtubule-binding proteins such as CRMP-2, APC, and microtubule- associated proteins (MAPs) such as MAP1b and Tau.87-90 Importantly, GSK3β inhibition not only leads to multiple axons but can also convert dendrites into axons.79 Of the downstream

MAPs, CRMP-2 overexpression is sufficient to induce multiple axons while expression of a truncated form impairs axon formation.90 Specifically, CRMP-2 links the Sra-1/WAVE1 complex to the plus-end orientated microtubule-based motor protein Kinesin 1, leading to actin dynamics at the tip of nascent axons.91 Similar results were observed with the other GSK3β targets APC, MAP1b, and Tau. Thus, once this particular break is removed (GSK3β inhibition),

18 selective transport through various kinesins can take place and enrich axon-specific proteins within the nascent axon.12, 68

1.4.2.3 - BDNF → TrkB → PKA → LKB1 → SAD-A/B/MARK-2 The Par proteins play a critical role in epithelial cell polarity and the Par3/Par6/aPKC complex is a well-known mediator of PI3K-dependent axon formation.92 Another Par protein,

LKB1 (or Par4), also lies upstream in this polarity pathway. Similar to PI3K, phosphorylated

LKB1 (pLKB1) is one of the few upstream proteins that has been labelled an axon determinant

(i.e. necessary and sufficient).68 Strong evidence by different groups demonstrated the requirement of LKB1 in axon specification during neuronal polarization in cultured cortical neurons as well as in vivo within the embryonic cortex.93-97 Local exposure of an undifferentiated neurite to brain-derived neurotrophic factor (BDNF) promotes axon specification that is dependent on PKA-mediated LKB1 phosphorylation on serine 431.95

Activated LKB1 in turn phosphorylates SAD-A/B and MARK-2 which reduces the affinity of these microtubule-binding proteins, thus destabilizing microtubules.93, 96 Here we have a well characterized pathway from an extra-cellular cue (BDNF) activating a series of kinases (TrkB →

PKA → LKB1 → SAD-A/B/MARK-2) converging on microtubules similar to the PI3K/AKT pathway. Interestingly, PI3K also gets activated by BDNF within the growth cone, suggesting parallel pathways being activated by the same extra-cellular cue.74, 98 Several studies have observed defects in axon formation in vivo but almost all of them displayed migration defects.68,

99-103 LKB1 knockdown or overexpression of a phospho-null mutant is one of a few examples of a defect in axon formation without the accompanied migration defect, highlighting the specific requirement of LKB1 in axon specification but not during radial migration into the cortical plate.96 In addition, this suggests that although there is great overlap with the molecular

19 mechanisms involved, radial migration can still take place normally even when the neuron cannot grow an axon.

1.4.2.4 - Ras superfamily of small GTPases Several members of the Ras- and Rho-family of small GTPases have been shown to regulate neuronal polarity.12, 104 Small GTPases are molecular switches that are active when bound to GTP. With their relatively slow intrinsic GTPase activity they remain active until their

GTPase activity is stimulated by interacting with GTPase activating proteins (GAPs).105 Thus,

GAPs act as negative regulators of small GTPase activity by promoting the GDP bound

(inactive) state. In contrast, they are activated by the exchange of GDP for GTP which is regulated by guanine nucleotide exchange factors (GEFs). The Ras-related protein 1B (RAP1B) was shown to act downstream of PI3K during neuronal polarization since an active RAP1B form was able to rescue axonal loss following PI3K inhibition.106 Furthermore, overexpression of

RAP1B induces multiple axon-like neurites while RAP1B knock-down with RNAi causes a complete loss of axons; a phenotype that is partially rescued by expression of an active form of

Cdc42.106 Thus, the accumulation of Rap1B at the tip of prospective axons follows PI3K activation and contributes to the recruitment of several downstream molecules involved in axon specification such as Cdc42.

Ras activation is coupled to several cell surface receptors, thereby relaying extracellular information to several downstream effectors such as PI3K and the canonical MAPK pathway. 107

Several members of the Ras-family of small GTPases have been shown to regulate neuronal polarity including H-Ras, R-Ras, K-Ras, and N-Ras.83, 106, 108, 109 Overexpression of these Ras members lead to production of multiple axons. The PI3K-mediated PIP3 enrichment induced by

Ras activation in turn leads to activation of Cdc42 and Rac1, two Rho family small GTPases well known for regulating the actin cytoskeleton.104

1.4.2.5 - Ras homologous (Rho) GTPases in neuronal polarization Ras homologous (Rho) family proteins are less than 25kDa in size and are members of the Ras superfamily of small GTPases.105 Among the 22 members of the Rho family of small

GTPases, Cdc42, Rac1, and RhoA have been the most extensively characterized.104, 105, 110 The great diversity of their regulators (82 GEFs, 67 GAPs, and 3 GDIs) highlights the diverse and tight regulation of these molecular switches in cell biology.104, 105 Expression of a cdc42 mutant

(Cdc42L28), which autonomously cycles between a GDP- and GTP-bound state, leads to multiple axons while neurons expressing a constitutively active Cdc42 fail to form any neurites.106 Similarly, loss of Cdc42 leads to severe defects in axon specification at least partially due to increased levels of phosphorylated cofilin (inactive), a regulator of actin dynamics enriched in developing axons.111 Taken together, this indicates that cycling between

GDP- and GTP-bound state is essential to normal GTPase function and the requirement of both

GAPs and GEFs are necessary to promote this cycling. Among the different GEFs involved in neuronal polarization, the Rac1-specific GEFs DOCK7, TIAM1, and STEF have been shown to be necessary to promote actin dynamics and microtubule stability.112, 113 Lastly, the RhoA- specific GAP, p190RhoGAP, is inactivated by GSK3β phosphorylation.114 As previously mentioned, GSK3β is inhibited during neuronal polarization by AKT phosphorylation downstream of PI3K.79, 86 This GSK3β inactivation in turn leads to increased p190RhoGAP activity and consequently a reduction in active RhoA-GTP.115 This highlights the negative role

RhoA plays during axon outgrowth. Mice lacking either of the two p190RhoGAP (A and B) genes display axon outgrowth defects.116 Taken together, Rho GTPase regulation is a crucial

21 intermediary step in cytoskeletal regulation allowing proper axon specification and outgrowth.

They are regulated by several GAPs and GEFs which remain poorly characterized. Since only a handful of GTPase regulators (GEFs and GAPs) have been described in the context of neuronal polarization so far and the fine-tuning of GTPase activity is necessary for proper physiological functions (ex: axon defect in Cdc42-deficient and constitutively active-Cdc42-expressing neurons); it will be interesting to see future studies addressing the role of other GAPs and GEFs.

1.4.2.6 - In vivo Unlike neuronal polarization in culture in a homogenous 2D environment containing only neurons, newly-born neurons are exposed to multiple signals and have to polarize and migrate along a specific trajectory. During cortical development, newly-generated projection neurons in the SVZ extend multiple neurites into the intermediate zone (IZ) similar to stage 2 cultured neurons (compare Figures 1.5 & 1.6).12, 117 These multipolar cells can migrate laterally before extending a trailing process that will become the axon and a leading process that will become one of several dendrites. It is these bipolar cells that migrate toward the cortical plate to develop into mature neurons.12, 117 This multipolar-to-bipolar morphology is a critical step in neuronal polarization in vivo and the formation of the trailing process (future axon) is the initial step, similar to cultured neurons (Figure 1.6). Several environmental cues have been shown to regulate this transition but a crucial step in regulating polarization in vivo is cell-cell interaction.118 A ‘Touch and Go’ model in which multipolar cells extend and retract immature neurites in the IZ suggests that once a neurite ‘touches’ the pioneering axons from early-born neurons, the neurite is stabilized and extend rapidly (Go).118 This ‘Touch’ mechanism is mediated by transient axonal glycoprotein-1 (TAG-1) expressed along pioneering axons while the ‘Go’ phase involves Lyn kinase and Rac1 activation.118 Several environmental cues such as

22 neurotrophins (NT-3, BDNF),95 Wnts,119 insulin-like growth factor (IGF-1),120 TGF-β,121

Semaphorin,122 Unc-6/netrin,123, 124 and Reelin,101 extracellular matrices, such as laminin,125 and

TAG-1118 promote axon specification in vitro and in vivo. The most compelling evidence for the role of an extracellular cue in regulating neuronal polarization and axon specification is that of

TGF-β.121 TGF-β2 is highly enriched in the VZ and forms a gradient along the radial axis where projection neurons migrate. High TGF-β2 within the VZ/SVZ and exogenous TGF-β2 is sufficient to direct the rapid growth and differentiation of an axon while cortical neurons lacking the type II TGF- β receptor fail to form any axons in vitro and in vivo. This study also showed that these TGF- β-dependent events were mediated by site-specific phosphorylation of Par6.121

Other than TGF- β, Semaphorin-3A secreted by superficial layers diffuses to form a gradient opposite to that of TGF-β.122 Sema3A and its receptors neuropilin-1 and plexins have been shown to play an important role in keeping bipolar neurons polarized toward the cortical plate.122

Another molecule forming a descending gradient, reelin, is secreted by Cajal-Retzius cells located in the marginal zone above the cortical plate and plays an important role in orienting multipolar cells towards the cortical plate.101 This orientation is mediated by activation of Rap1, which maintains N-cadherin levels at the cell surface. In addition, ephrin guidance cues and their Eph receptors were shown to play a similar role in regulating lateral migration of multipolar cells before radial migration of bipolar cells take place. Another classical axon guidance cue shown to regulate neuronal polarity is UNC-6/Netrin. UNC-6/UNC-40(Netrin/DCC) were shown to be required to break symmetry of HSC neurons in C. elegans while a recent study examined the role of netrin and its receptor DCC during cortical polarization in mice.124

Netrin/DCC signaling was shown to be involved in the multipolar-to-bipolar transition, thereby regulating both polarization and radial migration.123 Specifically, Netrin/DCC-mediated Fyn

23 activation led to tyrosine phosphorylation and activation of the adaptor protein Dab1. The best- studied pathway regulating cortical migration is the Reelin-Dab1 signaling cascade and this highlights an alternative means to activate Dab1 and promote cortical neuron polarization and migration. Taken together, several cues have been identified in the regulation of neuronal polarity and axon specification in vitro and in vivo. The intracellular mechanisms involved in vivo largely recapitulated what had been characterized in vitro.74, 117 A thorough description of all the players involved is well beyond the scope of this chapter and a brief selection of some well-defined pathways were mentioned here as examples. A common feature of these pathways is their convergence onto the microtubule and actin cytoskeleton. This convergence on the cytoskeleton drives the changes in cellular morphology necessary for neurons to polarize into their final and distinct polarized shape. In contrast to microtubules and the actin cytoskeleton, intermediate filaments (or neurofilaments) are relatively static and do not play any major role in the changes in morphology involved during neuronal polarization.

Figure 1.6: Polarization of cortical neurons in vivo

Newly-born neurons (blue) go through an initial multipolar morphology consisting of multiple neurites, similar to polarization in vitro. The critical multipolar-to-bipolar transition is defined by the emergence of a major process in the radial direction (leading process – LP). Radial translocation along radial glia (grey) leaves behind a trailing process (TP). This TP will become the axon (purple) that will grow into the intermediate zone (IZ) before exiting the cortex. Following the neurogenic period (E11~E17), dendrites (green) and axons terminally differentiate which include formation of the axon initial segment (yellow) and dendritic spine maturation. Adapted with permission from Annual Review of Neuroscience 2009 (Barnes & Polleux, 2009). Copyright © 2009 by Annual Reviews.

1.4.3 - Axon Outgrowth “It appears therefore, that the innumerable processes and intercellular connections offered by the adult nervous system can be interpreted as the morphologic expression of the infinite routes traced in space by currents of inducting or positive chemotropic substances during the entire developmental period. Thus the total arborization of a neuron represents the graphic history of conflicts suffered during its embryonic life.”10 Cajal, 1890 Neural circuits are the basis of neural function in health and disease. Investigating the molecular mechanisms involved in neural circuit assembly will help us understand neurodevelopmental disorders such as intellectual disability, autism spectrum disorders, and schizophrenia, enabling earlier diagnosis and develop a better therapeutic strategy. The molecular insights from these studies could also be exploited to improve therapies in the context of neural circuit disintegration such as Parkinson’s and Alzheimer’s disease. It is a long journey for the newly-born neuron in the VZ/SVZ to being part of the complex cortical neural circuitry and it all starts with neurite initiation.126 Polymerized actin structures such as filopodia and lamellipodia form nascent growth cones that protrude away from the cell body. There are two basic requirements for neurite formation: 1) dynamic peripheral actin network and 2) bundling of microtubules arrays.126 These dynamic cytoskeletal polymers provide the tips of growing neurites the ability to change shape and drive neurite outgrowth. As we have seen in the

25 previous section, intrinsic mechanisms are sufficient to initiate axon specification in one neurite.73 In the 3D environment of the embryo, combinations of signaling molecules help dictate the orientation of axon initiation before radial migration takes place (Figure 1.6). Much of the communication that occurs between environmental guidance cues and the cytoskeleton takes place at the growing tip of the axon, the neuronal growth cone (Figure 1.7). Once axon specification has begun, the growth cone takes charge of where the growing axon will end its journey to form functional synapses.

1.4.3.1 - The Growth Cone “The growth cone may be regarded as a sort of club or battering ram, endowed with exquisite chemical activity, with rapid amoeboid movements, and with certain impulsive force, thanks to which it is able to proceed forward and overcome obstacles met in its way, forcing cellular interstices until it arrives at its destination.”7 Cajal, 1890 The growth cone is a terminal expansion of growing axonal and dendritic tips.15 Under light microscopy, it can be divided into two distinct compartments: the peripheral and central regions (P and C regions)15 (Figure 1.7). Using high resolution imaging, a third region termed

‘transition zone’ (or T zone) was identified with distinct functions.127 As the name states it is the region between the periphery and central regions. The P region is a broad and relatively flat area characterized by dynamic filopodia and lamellipodia (Figure 1.7).15, 128 The C region, connected to the axon shaft, is enriched in organelles such as mitochondria and exocytotic vesicles (Figure

1.7).15, 128, 129 The majority of dense microtubule bundles terminate within the C region but single microtubules do venture out in the P region and interact with the actin cytoskeleton and cell signaling components to promote growth cone motility.15, 130, 131 Taken together, the growth cone can be considered as the vehicle driving axonal outgrowth while reading the road signs

(guidance cues) along the way with its cell surface receptors (Figures 1.7 & 1.8). The receptor- ligand interaction then triggers intracellular signaling cascades that will converge onto the

26 cytoskeleton to drive the proper turning response. Axon outgrowth is driven by the growth cone which goes through 3 different stages: 1) protrusion, 2) engorgement, and 3) consolidation.15, 128,

132 Protrusion is the addition of new membrane at the edge of the growth cone, driven by filamentous actin (F-actin) structures such as filopodia and lamellipodia. Microtubule-driven transport of organelles and vesicles into the peripheral region characterizes the engorgement stage. Finally, consolidation results from the contraction and stabilization of the central region into the cylindrical-shape axon shaft. Consolidation involves bidirectional movement of organelles and vesicles. Even when the axon is not growing, a constant combination of F-actin treadmilling and retrograde flow provides the growth cone motor the means to respond quickly to any changes in the environment.15, 128, 129 Retrograde flow is driven both by contractility of the motor protein myosin II in the T zone and polymerization in the P region. Overall the dynamic nature of the growth cone is driven by cyclical polymerization and depolymerisation of actin filaments. Thus, a thorough understanding of the molecular mechanisms involved in regulating the growth cone actin cytoskeleton is essential to understanding axon outgrowth and guidance.

Growth Cone Microtubule

Guidance receptor Axon Shaft C region G-actin lamellipodia F-actin

P region Mitochondria Exocytotic vesicle

Figure 1.7: The neuronal growth cone

The terminal expansion of growing axonal and dendritic tips is called the neuronal growth cone. The leading edge (P region) consists of dynamic, finger-like filopodia that explore the

‘road’. At the base of each protrusion, a mesh-like F-actin structure – lamellipodia, provide

structural support and connects to the transition zone (T zone). Although the majority of stable microtubules remain in the central region (C region), single exploratory (dynamic – pioneer) microtubule can be seen within the peripheral region and into filopodia. Actin rings

characterize the transition zone and ‘barrier’ that prevents organelles and vesicles from entering the peripheral region.

1.4.3.2 - Actin – the driver Actin filaments (F-actin), like microtubules, are polar polymers.133 The majority of actin monomers (G-actin) are added onto the barbed end and are bound to ATP while ADP-bound actin dissociates at the pointed end.133-136 Bundles of actin filaments are formed with the help of bundling proteins, leading to the formation of Cdc42-dependent filopodia superstructures that protrudes from the growth cone reminiscent of antennas sensing the environment whereas lamellipodia is characterized by a meshwork of F-actin at the growth cone leading edge and is dependent on the action of Rac1 (Figure 1.7).15, 104, 105, 128, 129 These actin superstructures, first characterized in Swiss 3T3 fibroblasts more than 25 years ago,110, 137 are not unique to neurons and are a signature of migrating cells.138 The clutch hypothesis proposed thirty years ago suggests that growth cone receptor binding to an adhesive substrate leads to the formation of a complex that acts like a molecular clutch, mechanically coupling the receptors and F-actin, thus anchoring Factin to prevent retrograde flow and driving actin-based protrusion of the growth cone on the adhesive substrate.139, 140 This interaction creates traction which is also dependent on myosin-II.141 Filopodia, the finger-like F-actin structures that protrude from the growth cone, act as guidance sensors at the front line of the growth cone and have a major role in establishing growth cone – substrate adhesive contacts during environmental exploration.15, 128 Actin

28 dynamics are necessary for directed axon outgrowth but overall dispensable for axon outgrowth.129, 142, 143 Neurons treated with agents that depolymerize F-actin are still capable of axon elongation but cannot initiate a turn in response to guidance cues.144, 145 A neuron without

F-actin can still grow but doesn’t know where it is going, reminiscent of a driverless car. Thus, the growing axon can be thought of as a car with actin as the driver, microtubules as the engine, the road is the various substrates and extracellular matrices it encounters while the various road signs are the secreted guidance cues (Figure 1.8). Interestingly, one group that specifically targeted filopodia structures found that these actin-based protrusions are indeed dispensable for accurate growth cone guidance but required for normal growth cone motility.146 It will be interesting to investigate at a super-resolution level the behavior of the growth cone when filopodia structures are eliminated. The balance between the rate of retrograde flow and polymerization determines if the growth cone extends or withdraws protrusions. If polymerization exceeds 3-6μm/min (retrograde flow rate), there is protrusion.134, 141 In addition, when cell-substrate adhesions increase, retrograde flow slows down and polymerization drives protrusions.134 Within the T zone, myosin-II plays an important role in severing actin filaments by exerting a contractile force on anti-parallel F-actin that contracts the actin meshwork and breaks it into small pieces.141 In terms of growth cone turning, it is widely assumed that repulsive guidance cues leads to disruption and loss of actin superstructures and actomyosin contraction while attractive cues leads to asymmetrical incorporation of actin on the side of the growth cone closest to the cue,15, 129, 134 although super-resolution cytoskeletal dynamics during growth cone turning is still lacking. The exquisite control of actin nucleation, polymerization/depolymerisation, bundling, and contraction involved in shaping the growth cone

29 is mediated by a complex repertoire of actin regulatory proteins which is beyond the scope of this introduction.129, 134

Long-rang Repulsion

Long-rang Attraction

Figure 1.8: The road signs throughout the road trip

The growth cone and its dynamic F-actin serve as the vehicle, driving directed-axon outgrowth,

reading the signs along the way making assuring the axon grows in the right direction. The adhesive substrate-bound cues such as cadherins and extracellular matrix proteins such as laminin and fibronectin pave the road (blue) while repellant substrate-bound cues such as ephrins act as guard rails (red), making sure the growth cone doesn’t deviate from its trajectory. In addition to the road (short-range/substrate-bound cues), several long-range cues (netrins & semaphorins) are

secreted within the embryo, forming gradients that growth cone can respond to by turning towards or away from (traffic lights). The driver (F-actin) has to remain vigilant and ready to respond to the several road signs. The constant F-actin treadmilling and retrograde flow provide the growth cone the necessary flexibility to respond in a timely manner, not that different to the driver with 30 his coffee to remain vigilant. The ‘Go’ versus ‘Stop’ signs does not derive from some intrinsic property of the guidance cue but rather depends on the receptors present at the membrane and consequently, the downstream signaling pathways activated.

1.4.3.3 - Microtubules – more than just an engine

Although microtubules have classically been considered secondary players in growth cone motility, axon outgrowth and growth cone turning requires the coordination of both the actin and microtubule cytoskeleton.130, 147-152 It was largely accepted that actin filaments initiated the change in direction (driver) while MTs followed their lead to promote outgrowth in that direction (engine).153, 154 As we will see, microtubules play a much more complex role than just driving outgrowth and can even steer the growth cone in response to guidance cues. Similar to actin polymers, MTs are polarized polymers consisting of α/β tubulin heterodimers held together in a head-to-tail fashion as their building block.133 Most MTs within the axon shaft are bundled together by lateral interaction of 11-15 protofilaments while some single MTs do explore and invade the peripheral region of the growth cone (Figure 1.7).147, 148 As mentioned above, axonal microtubules are homogenously oriented with their dynamic (+) ends towards the growth cone and away from the cell body.75, 128, 134, 147, 148 This orientation drives the rapid and increased rate of axonal versus dendritic outgrowth since MTs polymerization occurs mostly at the (+) end and dendrites have 50% of their MTs (+) end oriented toward the cell body. Highly dynamic growth cone MTs switch from growth and shortening phases called dynamic instability and function as a sensor to steer the growth cone in the right direction.134, 147, 148, 155 Furthermore, the dynamic pool of MTs are characterized by a post-translational modification in which a C-terminal tyrosine residue is added (tyrosinated MTs) that is required for directed outgrowth toward a target.156, 157 The importance of MTs in axon outgrowth and guidance had been highlighted by the many pathfinding defects found in humans caused by α- and β-tubulin mutation.158-160 An

31 increasing number of studies suggest that MTs play more than just a consolidation role for growth cone steering initiated by actin dynamics and play more of an instructive role than previously thought.148, 161, 162 Application of low concentrations of the microtubule-stabilizing drug taxol promotes polymerization at (+) ends and increases axon outgrowth in vitro and in vivo.163 Furthermore, disruption of MT dynamics in the growth cone completely abolished attractive and repulsive growth cone responses normally induced by diffusible cues such as

Netrin-1 and Sema3A, respectively. These two guidance cues, together with the other two classical guidance cues, Slit and Ephrin, and their signaling pathways will be introduced in the following section.155 It was shown that the local stabilization of MTs on one side on the growth cone induces attraction towards that side, highlighting the importance of MTs stabilization in growth cone turning. Interestingly, netrin-1-mediated growth attraction was shown to be mediated by the interaction of its receptor DCC with a β-tubulin isoform through direct TUBB3-

DCC interaction leading to MTs stabilization.164 This interaction is dependent on Src family kinase phosphorylation of TUBB3 which was also shown to phosphorylate DCC in response to netrin-1 stimulation.165, 166 Lastly, TUBB3 mutations partially phenocopies the midline crossing defects observed in both DCC and Netrin loss-of-function, highlighting its role in this signaling pathway.158-160 Taken together, the dynamic nature of growth cone microtubules and growth cone morphology makes it hard to visualize in real-time.134 Novel super-resolution techniques will definitely add to our detailed understanding of microtubule-mediated growth cone motility.

1.4.3.4 - Actin-Microtubule interaction Coordination of the two dynamic cytoskeletal polymers need to be tightly controlled to assure the growth cone responds to the environment in a coordinated manner. Three modes of coordination between actin filaments and MTs have been documented.130 First, actin

32 polymerization and actomyosin contractibility influences MTs dynamics.127, 130 As previously mentioned, actin retrograde flow restricts the entry but can also lead to the direct removal of

MTs within the P domain and following filopodial F-actin bundle contraction and depolymerisation, actin-bound MTs can be transported rearward. In addition, contractile actomyosin arcs exert a pulling force that facilitates MT bundling and advancement into the C domain. Second, filopodia act as guides for single MTs from the C domain exploring the periphery.127, 130, 167 Third, actin-mediated MT capture is thought to be important for the dynamic regulation of adhesion sites following coupling of cell-adhesion molecules with actin.168

Other than these three modes of coordination, several actin/MT cross-linking proteins have been identified in the context of axon outgrowth and guidance. The +TIP protein spectraplakin has been shown to bind MT lattices via its GAS2 domain while binding actin with its CH domain.130,

169 Deleting just one of the two mammalian spectraplakin, ACF7, in mouse cortical neurons led to unstable MTs leading to defects in leading process morphology and cortical migration.170

Dynein was shown to target MTs to the peripheral actin network of neuronal growth cones through the formation of a Lis-dynactin-dynein-EB3 complex which is important for growth cone motility.171 Finally, actin itself can transport MTs in the anterograde direction to regions of low MT density.172

1.4.3.5 - Intermediate filaments (Neurofilaments) Although neurofilaments are an important structural component of the neuron’s cytoskeleton,173 they are largely dispensable during axon outgrowth and guidance due to their static nature.174 If they do play an important role, it remains largely unknown.175

1.4.3.6 - Ras GTPases

Ras GTPases are activated by a large number of plasma membrane growth factor receptors and adhesion receptors leading to the activation of the MAPK and PI3K signaling pathways, among others.104 Less is known about the role of the Ras family members in axon outgrowth compared to Rho GTPases although Ras was shown to be involved in NGF-mediated axon outgrowth through activation of Raf.176 In addition, Ras inactivation downstream of semaphorin signaling is involved in growth cone collapse.104, 177, 178 Ras has also been shown to play a role in Eph/ephrin-matrix adhesion179 and netrin/DCC-mediated axon outgrowth and guidance.180

1.4.3.7 - Rho GTPases Introduction of the mechanisms involved in regulating the actin and MT cytoskeleton wouldn’t be complete without addressing the critical role of the small Rho GTPases.104, 105 Their ability to coordinate multiple signal pathways with the help of their regulators (152 total) with precise spatial and temporal control drives most of the activities that underlie axon outgrowth and guidance.104 The molecular cycle regulating cell motility was first proposed in migrating fibroblasts in a two-dimensional environment.110, 137 First, there is protrusion at the leading edge, followed by adhesion of the leading edge (and de-adhesion of the trailing edge) and finally contraction of the rear that pushes the cell body forward.138 In this model, the cycling activity of

Rho GTPases is responsible for the protrusion-retraction cycle. Rac1 and Cdc42 promote actin polymerization and protrusions at the leading edge by stimulating the actin nucleation factor

Arp2/3 through the activation of the Wiskott-Aldrich syndrome protein (WASP) family members, WASP-family verprolin-homologous protein (WAVE) and neural WASP (N-WASP), respectively.105, 138 While Rac1 activity generates lamellipodia via the regulation of WAVE,

Cdc42 activity induces the formation of filopodia via N-WASP. In contrast, RhoA/Rho kinase

(ROCK) activity at the trailing edge promotes contractibility by activating myosin light chain.105,

138 In the context of axon outgrowth much of the same mechanisms apply to growth cone motility.104 Briefly, RhoA can either promote or inhibit axon extension depending on the type of effector: mDia or ROCK, respectively.104 Kalirin-9 (RhoGEF) activates while p190RhoGAP inactivates RhoA to promote axon outgrowth.104, 181 Both ROCK and PAK (Rac1 effector) can inhibit cofilin through LIMK.128, 134, 136 The balance of dephosphorylated (active) and phosphorylated (inactive) cofilin appears to be crucial for axon extension. Several GAPs (ex: α- chimerin, p190RhoGAP, & p120RasGAP)180-182 and GEFs (ex: Tiam1, STEF, Dock180, &

Trio)112, 183-185 have been documented to be involved in axon outgrowth and guidance and we will see where some of them come into play in the context of the 4 classical guidance cues and their signaling pathways.

1.4.4 - Axon Guidance – The Road Signs It appears that with this hypothesis we have shed light into a dark cave, when in reality we have explored only the entrance, from which its imposing abyss appears even more distant and black. On what bases are mechanical influences guiding the created ameboid streams? Which is the cause of certain preferences of time and location in the distribution of secretory phases? Why does the chemotactic sensitivity cease of decrease in certain periods? These are questions that present day Science can only pose: their clarification, i.e., their total reduction to physicochemical mechanism, will be the work of the future.7 Cajal, 1890 The development of neural circuits comprising billions of neurons interconnected with trillions of synapses17 depend on the correct pathfinding of axons to their final target. Today, most if not all, guidance molecules and receptors have been identified and in comparison to the complexity of neural circuits, the number is surprisingly small.186-188 Similar to the multiple stops we take during a long road trip, axons navigate through a series of intermediate steps along the way to their final destination, their future synaptic partners. As highlighted by Cajal well over a century ago, how do neurons switch their chemotactic response at different intermediate steps? If growth cones are attracted to the intermediate target, why don’t they form synapses within it? We now know that receptors on the cell surface of growth cones mediate the response

35 to guidance molecules present throughout the embryo and that instead of switching the guidance molecules present at different intermediate steps, which would require that all axons travel at the same time, neurons change the receptor population present at the cell surface.186-188 Switching the response quickly within the intermediate target prevents staling and ensures the axon reaches its final target in a timely manner.189 It was demonstrated that the correct navigation of axons depend on a combination of long and short-range attraction and repulsion (Figure 1.8).190 Long range guidance molecules form gradients that can attract or repel axons far from its source while short-range are mostly contact-dependent (Figure 1.8). All four classic guidance cues first identified in the context of commissural axon guidance in the spinal cord have also been showed to be involved in cortical axon guidance.

1.4.4.1 - Classical guidance cues and commissural axon guidance The best understood model for axon guidance is the dI1 interneurons of the dorsal spinal cord.188, 191 Interneurons located on the dorsal side of the developing spinal cord, dI1 neurons, first send their axons ventrally toward their intermediate target, the floor plate, cross the midline before turning rostrally (Figure 1.9). The midline crossing of these dI1 interneurons create commissures along the spinal cord and are thus often called commissural neurons. This model for studying axon guidance represents a convenient model as the initial trajectory is easy to follow experimentally.24 The spinal cord, along with the retina, are also two regions of the CNS with an easier accessibility for manipulation (ex: open-book preparation). It was in these commissural axons that the four major family of guidance cues were first identified: Netrins,

Slits, Semaphorins, and Ephrins.188, 190-192 Other factors first identified in a different context have also been identified in the spinal cord as having guidance properties.186, 192, 193 Briefly, dI1 neurons born on the dorsal side of the spinal cord are attracted to gradients of netrin-1 secreted

36 from the the floor plate (Figure 1.9).194 This attraction to the midline is mediated by the netrin-1 receptor DCC expressed at the cell surface of growth cones.195, 196 Upregulation of the Slit receptor Robo1 within the floor plate switches the response from attraction to repulsion by favoring the Slit-Robo-mediated repulsion (Figure 1.9).197 In addition, interaction between

Robo1 and DCC attenuates the netrin/DCC-mediated attractive response.198 Class-III

Semaphorins derived from the floor plate provide a repulsive signal for post-crossing axons expressing Neuropilin-2 and PlexinA1 (Figure 1.9).199 This particular repulsion is prevented in pre-crossing neurons by calpain-1-mediated PlexinA1 cleavage.199 The role of calpain-1 in axon guidance in not limited to Semaphorin/Plexin signaling and chapter 3 will demonstrate that it plays an important role in netrin/DCC-mediated axon outgrowth. Finally, Ephrin-B3 expressed in the floor plate also mediates repulsion away from the midline through interaction with its receptor Eph-B3.200 The complexity of neural circuits compared to the relatively small number of guidance molecules and receptors is achieved, at least in part, through the crosstalk of these signaling pathways, thereby increasing the diversity of cellular responses to the limited number of cues.187, 188, 201 Since our lab focuses on cortical axon outgrowth and guidance, I will introduce these four guidance cues and their role in cortical axon guidance after taking a brief look at other molecules first identified in another context that were shown to have guidance properties.

1.4.4.2 - Morphogens and growth factors in commissural axon guidance Some of the best-known morphogens have also been found to have guidance properties, independent of their role in providing positional information to initiate a differentiation cascade mediated by specific combinations of transcription factors.48, 49, 193 Sonic hedgehog (Shh) secreted from the notochord and eventually the floor plate of the spinal cord forms a ventral-high

37 to dorsal-low gradient, providing neural progenitors positional information required to activate a specific series of transcription factors required for differentiation into a specific neural subtype.202 Although Shh plays a similar function as netrin-1 in attracting commissural axons toward the midline, its effect is weaker than that of netrin-1 as the effect can only be detected in embryos defective in netrin-1 signaling (Figure 1.9).203 Similarly, vascular endothelial growth factor (VEGF), also act as a chemoattractant to guide commissural axons toward the floor plate

(Figure 1.9).204 Interestingly, all three chemoattractant (netrin-1, Shh, and VEGF) converge on

Src family kinases to mediate their chemoattractive properties.165, 166, 204-206 In addition, Shh expressed in a caudal-high to rostral-low gradient acts as a repellent in post-crossing commissural axons, guiding them rostrally toward the brain (Figure 1.9).207 Conversely, BMP7 and Draxin form an opposing gradient coming from the roof plate that repel commissural axons away from the roof plate and consequently toward the floor plate (Figure 1.9).208, 209

Furthermore, Wnt4 expressed at the ventral midline in a rostral-high to caudal-low gradient attracts post-crossing commissural axons rostrally (Figure 1.9).210 Finally, Wnt1 and Wnt5 have been implicated in chemorepulsion of the CST and corpus callosum.211

Pre-crossing Post-crossing

- - BMP7 - - BMP7 Draxin Draxin ------Anterior - -

+ + Netrin-1 - - - + + VEGF - - - + Netrin-1 Shh Shh ephrin - - Sema VEGF Slit

Figure 1.9: Commissural axon guidance in the spinal cord DI1 neurons born on the dorsal side of the developing spinal cord first send their axons ventrally toward the floor plate, cross the midline before turning rostrally. Dl1 axons are

attracted to the midline by a combination of floor plate-secreted netrin-1, Shh, and VEGF and repelled by roof plate-secreted BMP7 and Draxin. By regulating receptors present at the cell surface, the attractive netrin-1 and VEGF response is attenuated in favor of the repulsive response mediated by Slit, Sema, and ephrin while the initial attraction mediated by Shh is converted to repulsion. Post-crossing axons are attracted rostrally toward a Wnt4 gradient and

away from a Shh gradient. 1.4.4.3 - Classical guidance cues 1.4.4.3.1 - Slit/Robo signaling at the roundabout The highly conserved family of secreted repulsive guidance cues, Slits, were first identified through a genetic screen in Drosophila and found to be secreted at the ventral cord midline to repel ipsilateral projecting neurons and post-crossing commissural axons.212-214 Three

Slit genes have been identified in vertebrates encoding ~200kDa proteins. They consist of four stretches of leucine-rich repeat (LRR) domains (D1-D4), seven to nine epidermal growth factor

(EGF) repeats, an Agrin-Perlecan-Laminin-Slit (ALPS)/Laminin-G-like domain, and a C- terminal cysteine knot.197 The three Slit receptor genes, Robo, were also identified in

Drosophila.215, 216 Their name was coined due to the phenotype of Drosophila mutants in which commissural axons stalled at the midline of the ventral nerve cord, creating an axonal

‘ROundaBOut.217 Four mammalian orthologues exists with Robo1-3 containing five immunoglobulin-like domains and three fibronectin type-3 repeats within its extracellular domain, a single-pass transmembrane domain and a cytoplasmic tail containing several conserved regions termed CC (cytoplasmic conserved) domains.197 These domains, however, are quite variable between species. Robo1/2 are well-conserved with overlapping functions during axon guidance while Robo3/4 do not actually bind Slits.197, 218 Robo3 in particular was shown to be involved in opposing the repulsive response mediated by Slit. Interestingly, Robo3 also plays

39 a positive role in netrin/DCC-mediated growth cone attraction.218 Upon Slit/Robo binding through Slit’s LRR2 and the first Ig domain of Robo, the cytoplasmic tyrosine kinase Abelson play an important role by phosphorylating Robo’s CC1 domain as well as downstream effectors required for growth cone repulsion.187, 197, 219 Downstream, the activation of Rac1 and RhoA, together with the inhibition of Cdc42 mediate the required cytoskeletal rearrangements.220 The adaptor protein Nck, which binds Robo receptors, was shown to mediate the link between Robo receptors and GTPase activation to mediate cortical axon outgrowth and branching.221 In

Drosophila, specific Rac GEFs and GAPs linking Robo receptors directly to the GTPases have been identified.220, 222 The Slit-Robo-GTPase activating proteins (srGAPs) Vilse/SrGAP was shown to play an important role in growth cone repulsion in a dose-dependent manner.223, 224

Too much or too little Vilse/SrGAP lead to defects in growth cone repulsion at the midline.

Thus, the fine-tuning of Rac activity seems to be required to mediate midline repulsion and avoid either stalling or re-crossing at the midline.

One of the great enigmas in the field of axon guidance was the mechanism by which growth cones switch their attractive response to the midline to a repulsive one. The predicted consequence of midline attraction was either growth cone stalling within the floor plate or continuing re-crossing after exiting the midline. If Slits were expressed in the floor plate before axons crossed; why were the growth cones unresponsive to the repulsion that kept post-crossing axons away from the midline? The initial Drosophila screen that identified the Robo receptors also identified another mutant with severe midline crossing defects, commissureless (Comm).225

Comm, specifically expressed in pre-crossing commissural axon (and not in ipsilateral-projecting axons) is a transmembrane receptor that targets Robo to the lysosome for degradation.226 This maintains low levels of Robo receptors in pre-crossing axons and consequently prevents

40 repulsion. Following midline crossing, Comm expression is downregulated to enable Robo- mediated repulsion out of the midline and prevents re-crossing.197, 226 No vertebrate homologue of Comm has been identified and alternative mechanisms have evolved to compensate to regulate midline crossing. As we will see below it involves the crosstalk between different signaling pathways.

1.4.4.3.2 - Semaphorin/Plexin signaling The more than 20 semaphorin proteins identified thus far are categorised into eight subfamily classes (Semaphorin Nomenclature Committee, 1999).227 Invertebrate semaphorins are grouped into classes 1, 2, and 5 while classes 3-7 represent vertebrate’s semaphorins (except for

Sema-5C only found in vertebrates).227, 228 All semaphorins are defined by a ~400 amino acid sema domain which facilitates interaction with their receptors as well as mediating homodimerization. Their principal receptors, plexins, are transmembrane proteins divided into four classes (A-D).229, 230 Two invertebrate plexins, PlexA and PlexB, and nine plexins have been identified in vertebrates. Unlike other guidance receptors, plexins contain a catalytically active intracellular domain, a GTPase-activating (GAP).229, 230 Among the other receptors/co- receptors identified, class-3 semaphorins (except Sema3E) which cannot bind direclty to plexins require the high affinity binding of neuropilins as co-receptors.230 Two neuropilins have been identified in vertebrates: neuropilin-1 and -2 (Nrp1/2). Both Nrp1 and Nrp2, together with plexins, form a holoreceptor complex in which they are obligate ligand-binding subunit of class

3 semaphorins.228 Furthermore, Sema7A can signal independently of plexins through β1- integrins whereas Sema3A can also signal through Nrp1/L1cam receptor complex. Upon semaphoring binding, the GAP domain of plexin gets activated to reduce Ras and Rap activity in neurons.228 This reduction in active Ras, in turn, leads to activation of the MT-depolymerizing

41 protein, CRMP-2, via PI3K-AKT-GSK3β. Binding of the small GTPase Rnd1 to plexin B1 was shown to stimulate its GAP activity while PKA-mediated GAP domain phosphorylation triggers

14-3-3ε recruitment and a concomitant reduction in GAP activity.178, 231 In Drosophila, direct regulation of the actin cytoskeleton by plexins is accomplished via the Mical family of redox enzymes.232 These enzymes can bind plexins and mediate F-actin subunit oxidation and disassembly. In addition, Sema3A regulate actin dynamics through the phosphorylation of cofilin by LIMK.233 Although widely recognized as repellents, semaphorins can also mediate chemoattraction in the CNS through expression of various receptors/co-receptors.228 The presence of Nrp1 and VEGR2 on subicular axons in the brain switches Sema3E-mediated repulsion to attraction through VEGFR2-mediated downstream signaling.234 Interestingly, semaphorins and plexins can also engage in cis-interactions to prevent interactions in trans and/or activate plexin signaling in cis.228 Taken together, this illustrates the versatility of semaphorins guidance cues during axon guidance.

1.4.4.3.3 - Ephrins/Eph signaling A considerable backlog of supporting observations suggestive of chemotaxis has accumulated during the last two decades, but has not been emphasized owing to the prevailing unpopularity of anything suggestive of neurotropism and chemical guidance. With the demonstration of chemotactic regrowth in the optic system leaving little room for further doubt of the selectivity effects, the way seems cleared for retrospective reassessment of some of the earlier observations with an increased recognition of the importance of selective chemoaffinity in nerve growth and connection in general. R. W. SPERRY, 1963235 It was ephrin/Eph signaling in the visual system that inspired Roger Sperry’s chemoaffinity theory based on pioneering studies in the newt’s retinotectal system.235, 236 The discovery of Eph/ephrin signaling came over thirty years later in the context of topographical mapping of the visual system.237, 238 Since then, Eph/ephrin signaling have been implicated in guiding axons of the auditory, somatosensory, and olfactory systems, as well as playing a critical role in midline crossing of spinal commissural and cortical callosal axons.200, 239-243 They are the

42 largest family of receptor tyrosine kinase (RTK) encoded in the mammalian genome.243, 244 The receptors are divided into two subclasses, EphA (EphA1-EphA8) which bind GPI-linked ephrin-

As (ephrinA1-ephrinA5), and EphB receptors (EphB1-EphB6), which bind transmembrane ephrin-Bs(ephrinB1-ephrinB3). Ephrins are short range guidance cues since they are attached to the membrane.243, 244 One unique feature of ephrin/Eph signaling is that it can occur in both directions; forward signaling is transduced by the receptor, Eph, while reverse signaling happens downstream of the ephrin ligands.245 Although widely known as a repellant, both forward and reverse signaling can repel or attract the growth cone.243, 244 As with the other guidance receptors, ephrin signaling influences the activation state of small Rho GTPases. Ephrin-A5 signaling activates RhoA and inactivates Rac1/Cdc42 through the constitutively-bound RhoGEF, ephexin, leading to growth cone collapse.246 In addition, Eph receptor activation leads to the inhibition of Ras and the MAPK pathway.247, 248 Direct link of Eph receptors to the cytoskeleton has been shown through the association and dephosphorylation of Abl upon ephrin binding.249

Reverse signaling through ephrins on the other hand is initiated by Src-mediated tyrosine phosphorylation of ephrins that create binding sites for the SH2/SH3 adaptor protein Grb4.

Grb4, in turn, regulates FAK activity and the redistribution of paxilin.249 Ephrin/Eph signaling play an important role in mediating adhesion (Figure 1.8). Binding of the phosphotyrosine phosphatase Shp-2 to EphA2 leads to FAK inactivation and decreased intergrin-mediated adhesion whereas increased adhesion is induced by ephrin-A via Src kinases.250 Crosstalk between Eph/ephrin and Semaphorin/Nrp1 during midline crossing has recenlty been demonstrated (see below).242

1.4.4.3.4 - Netrin - the one who guides Direct support for Cajal’s proposed chemotropic guidance of growth cones was only obtained a century after he published his chemotactic hypothesis with the identification of the netrin family of guidance cues.194, 251, 252 Netrin was first discovered as a secreted molecule capable of guiding C.elegans axons towards Unc-6-secreting epidermoblasts located at the midline.252 It was 2 years later that the verterbrate homologue of Unc-6 was identified and named netrin based on the Sanskrit word “netr” meaning “the one who guides”.194, 251 It was purified based on its ability to attract commissural axons of the spinal cord towards the floor plate. Similarly, the C. Elegans orthologue of DCC, Unc-40, was first identified in the same genetic pathway as Unc-6 in C. Elegans.253 It was later characterized as an Unc-6 receptor in parallel to the characterization of DCC as a netrin receptor.195 These pioneering studies paved the way for the identification of other guidance cues and the field of axon guidance in general.

Vertebrates express 4 different secreted netrins (netrin 1-4) and two related GPI-anchored membrane proteins, netrin-G1 and netrin-G2.196 All netrins are composed of ~600 amino acids and belong to the superfamily of laminin-related proteins composed of an N-terminal laminin VI domain, followed by a series of EGF-like domains and a C-terminal netrin-like domain (NTR)

(Figure 1.10).196 Netrins 1-3 are highly similar and are all capable of mimicking the chemoattraction of the floor plate. The best-studied member of the family is netrin-1. Netrin receptors include DCC, the DCC paralogue neogenin, UNC5A-D, the G protein-coupled

Adenosine 2B receptor (A2b), and Down syndrome cell adhesion molecule (DSCAM). Netrins are bidirectional cues; they promote attraction when bound to DCC and repulsion when bound to

DCC/UNC5. 196, 254

Figure 1.10: Structural domains of netrin Netrins are members of the laminin superfamily. All netrins are ~ 600 amino acids with their N-terminal laminin VI domain, followed by a series of EGF-like domains and 44a C -

terminal netrin-like domain (NTR).

1.4.4.3.5 - Netrin - the one who guides... …locally Netrin was thought to attract commissural axons by forming a gradient with the highest concentration close to its source, the floor plate. However, 20 years later, two independent laboratories reported that netrin-1 secreted from the floor plate is dispensable for commissural axons to reach the floor plate.255, 256 Netrin secreted locally from precursor cells in the ventricular zone of the spinal cord is necessary and sufficient to guide commissural axons.

Importantly, deletion of netrin specifically in the VZ phenocopies axon defects previously described in netrin-1 knockout mice.255, 256 Thus, although netrin-1 does form a gradient along the dorsal-ventral axis of the spinal cord, it appears that at least for commissural axons, netrin-1 functions as a short-range contact-dependent attractive cue by promoting growth cone adhesion.

1.4.5 - DCC - The One Who Attracts 1.4.5.1 - Structure

DCC was first identified as a candidate for a tumor-suppressor gene on human chromosome 18q.257 It was later characterized as the key receptor mediating the attractive response to netrin.195 Inactivation of the DCC gene in mice causes severe defects in axonal projections, similar to netrin-1-deficient mice.251, 258 DCC is a type I transmembrane protein consisting of four immunoglobulin-like domains (Ig) and six fibronectin type III domain (FN3) on its extracellular domain and three conserved P domains (P1, P2, & P3) on the intracellular side (Figure 1.11).195, 196, 259 The structure of the cytoplasmic tail of DCC remains largely unknown while its extracellular domain has been resolved alone and in complex with netrin-1.260,

261 The four N-terminal Ig-like domains of DCC fold into a horseshow-like conformation

(Figure 1.11).262 This unique horseshoe conformation turns out to be a conserved structural feature in many neuronal receptors such as another netrin-1 receptor, DSCAM and the adhesion receptor TAG-1.262 Several binding sites have been identified by two independent laboratories and the FN5 domain of DCC and the EGF domains of netrin seem to be crucial for downstream signaling (Figure 1.12).259-261, 263 This highly specific first binding site involves netrin’s EGF3 domain and DCC’s FN5 domain. A second binding site between netrin’s EGF1 and EGF2 domains and DCC’s FN5 and FN6 was resolved.261 Finally, netrin’s laminin domain was found to interact with DCC’s FN4 domain.260 Thus, netrin-1 binds DCC mainly through DCC’s FN5 domain and to a lesser extent with FN4 and FN6 domain. Importantly, netrin-1 always associates with two DCC molecules, confirming previous findings (Figure 1.12).264 Importantly, in its dimerized form one DCC molecule never physically interacts with the other DCC molecule within the extracellular domain. The closest distance between the two DCC molecules is

20Ǻ.260, 261, 263 Physical interaction could be possible via its intracellular P3 domain as this domain is required for self-association (Figure 1.11).

Ig2 Ig1

DCC

FAK Ig3 Ig4 FN1 FN2 FN3 FN4 FN5 FN6 P1 P2 P3 TM

Figure 1.11: DCC is a type I transmembrane receptor part of the immunoglobulin superfamily

The extracellular domain consists of four N-terminal Ig domains folded in a horseshoe-like conformation followed by six fibronectin type-III domains. A short transmembrane domain links the intracellular tail consisting of 3 conserved P domains responsible for downstream signaling

g 2 3 Ig2

Ig3 Ig3 Ig2

I I

g g

1 4

2 Ig1 Ig4 Ig4 Ig1

F F

F N

3 3

Netrin 3

N N EGF1 EGF2 EGF3

VI NTR N 4

t 5

6 6 6

P 1 P1 P1

P 2 P2 P2

K P

FAK A P3 P3 3 F

Figure 1.12: Netrin-1 associates with two DCC molecules

The first and highly specific binding site occurs between netrin’s EGF3 domain and DCC’s FN5

domains. Netrin’s EGF1 and EGF2 also bind FN5 and FN6 while netrin’s laminin domain can

bind DCC’s FN4 domain. The close proximity between the two P3 domains of each DCC

molecules likely mediates dimerization on the intracellular side.

1.4.5.2 - Signaling DCC plays two important roles in netrin-1-mediated axon outgrowth and attraction: 1) it serves as a transmembrane bridge linking extracellular netrin to the actin cytoskeleton and 2) as a signaling platform promoting actin dynamics and activating several downstream effectors.196 In the absence of netrin-1 DCC is constitutively bound to the adaptor protein Nck1 and the focal adhesion kinase (FAK).166, 265-267 Netrin-1 binding triggers DCC homodimerization via its P3 domain (Figure 1.11), autophosphorylation of FAK, as well as the recruitment and activation of

Src family kinases.166, 206, 265 Netrin-1-mediated traction force and FAK activation necessary for

47 axon outgrowth were also shown to be dependent on myosin II.268 Surprisingly, activated Src kinases lead to Robo3 phosphorylation and activation that promote DCC interaction.218

Interestingly, Robo3 does not mediate Slit-mediated repulsion and only weakly binds to Slit.

Instead, Robo3 has been repurposed during mammalian evolution to promote netrin/DCC- mediated attraction instead of Slit-mediated repulsion.218 Our lab has shown that activation of

Fyn and the resulting Fyn-mediated tyrosine 1418 phosphorylation of DCC is crucial for downstream signaling and the overall growth cone response to netrin/DCC.206 This led us to investigate one particular SH2-containing protein associating with pY1418, p120RasGAP, that will be discussed in chapter 2. Although phosphorylation on tyr 1418 is required for signaling, most phosphorylation on DCC occurs on ser/thr residues but their function has never been examined.206 Ser/thr phosphorylation on DCC’s cytoplasmic tail will be discussed in chapter 4.

In addition to DCC phosphorylation, activation of Fyn leads to the activation of Rac1 and Cdc42 and inhibition of RhoA.183, 269-271 Two Rac1-GEFs, Trio and DOCK180, promote Rac1 activation downstream of DCC activation.183-185 Interestingly, several netrin-dependent projections were affected in Trio knockout mice. Projections that form the internal capsule and corpus callosum were defective while the anterior commissure was completely absent.183 In addition, Trio-mediated Rac1 activation downstream of netrin/DCC was shown to be dependent on Hsc70 chaperone activity.272 Downstream of Rac1 activation, p21(Rac1)-activated kinase-1

(Pak1) gets activated by phosphorylation and recruited to DCC-bound Nck1.269 Downstream of the other activated GTPase, Cdc42, Ena/Vasp and N-WASP are activated and induces the formation of filopodia protrusions.269 In parallel, Src kinases and RhoA/Rho kinase-mediated activation of ERM proteins downstream of netrin/DCC signaling leads to an increase in ERM-

DCC located within filopodia.273 ERM proteins in turn mediate a direct physical interaction of

PKA with DCC, leading to PKA activation.274 Activated-PKA can then phosphorylate and activate Ena/Vasp proteins. Under basal conditions, Vasp is ubiquitinated by the E3 ubiquitin ligase TRIM9 and this reduces its filopodia tip localization, Vasp dynamics, and filopodia stability.275 Upon netrin-1 stimulation, Ena/Vasp gets deubiquitinated, thus promoting its tip localization and filopodial stability.275 Interestingly, the same group recently found that TRIM9- mediated ubiquitination of DCC blocks interaction and activation of FAK.276 Upon netrin-1 stimulation, this ubiquitination is reduced, thus promoting FAK interaction and activation.276

Interestingly, in addition to phosphorylating and activating Vasp, a novel Robo1 coreceptor,

FLRT3, was shown to be required for netrin/DCC-mediated attraction by increasing the cell surface pool of DCC through activation of PKA downstream of Slit1/Robo1/FLRT3 signaling.277

This crosstalk between netrin/DCC and Slit/Robo signaling is not unique and highlights the direction in which the field of axon guidance is going.187, 188, 201

Netrin/DCC signaling also activates the MAPK pathway leading to extracellular signal- regulated kinases 1 and 2 (ERK1(MAPK3) and ERK2(MAPK1) phosphorylation and recruitment to the DCC receptor complex.278, 279 This ERK1/2 activation was shown to be required for netrin-1-induced protein synthesis (see below).280 Netrin/DCC signaling also promotes the synthesis of phosphoinositide phosphatidylinositol (4,5) biphosphate (PIP2) that is phosphorylated by PI3K to produce PIP3.281, 282 Just as PIP3 helps the recruitment of AKT to the plasma membrane to promote neuronal polarization and axon specification, this particular PIP3 enrichment helps recruit Rho GTPases to the plasma membrane where it regulates the actin cytoskeleton.104, 105 Conversely, netrin-1 induces PIP2 hydrolysis by phospholipase Cγ (PLCγ) to generate diacylglycerol (DAG) that activates protein kinase C (PKC) and inositol 1,4,5-

2+ 281, 282 triphosphate (IP3) that stimulates the release of intracellular Ca . The calcium release is

49 required to induce netrin-dependent growth cone turning. Intracellular calcium concentration is also increased via the activation of transient receptor channels (TRPC) that allows Ca2+ influx from outside the cell.283, 284 Calcium signaling serves many cellular functions such as changing a protein’s conformation upon calcium binding.285 In particular, the calcium-dependent cysteine protease, calpain, and its activation downstream of netrin/DCC signaling will be discussed in chapter 3.

1.4.5.3 - Netrin/DCC and the cytoskeleton In addition to regulating the cytoskeleton indirectly through the various signaling effectors mentioned above, netrin-1 regulates the translation of β-actin to promote growth cone turning.286, 287 Specifically, netrin-1 activates a translation initiation regulator, eIF-4E-binding protein 1 (4EBP) asymmetrically, leading to an increase in β-actin translation on the side of turning. The local transport and translation of β-actin mRNA downstream of netrin-1 stimulation is accomplished with the help of zipcode binding protein 1 (ZBP1), an RNA-binding protein which binds a specific sequence on the 3’UTR of β-actin mRNA.286 This netrin-1-induced β- actin mRNA translation is likely mediated by DCC’s direct physical interaction with several components of the translation machinery such as eIFs, ribosomal subunits, and monosomes.288

The involvement of DCC in netrin-1-induced β-actin was also demonstrated indirectly by the fact that ERK1/2 activation, which depends on netrin/DCC signaling, is required for netrin- induced protein synthesis.289 In addition to promoting β-actin translation, DCC interacts with

MTs downstream of netrin-1 stimulation.164 It was shown that DCC interacts directly with the most dynamic tubulin isoform in neurons, βIII-tubulin (TUBB3), in a netrin-1-dependent manner to promote MT dynamics. Reducing TUBB3 levels strongly impaired netrin/DCC-dependent projections.164 Importantly, mutations in TUBB3 found in humans strongly affected netrin-

1/DCC signaling and mice harboring these mutations phenocopied the projection defects observed in netrin-1 and DCC knockout mice.160

1.4.5.4 - DCC in human axon guidance DCC mutant mice exhibit a failure of the CST to cross the midline and a complete agenesis of the corpus callosum which is characterized by the absence of the corpus callosum.195,

290 Similarly, germline DCC mutations in humans cause midline crossing defects that leads to abnormal corticospinal track (CST) and corpus callosum development.291-294 Interestingly, monoallelic DCC mutations cause congenital mirror movements (CMM) associated with abnormal midline crossing of the CST and/or agenesis of the corpus callosum (ACC).293

Congenital mirror movements are pathological involuntary movements that mirror intentional movements on the contralateral side. Biallelic DCC mutations on the other hand cause a more complex syndrome; developmental split brain syndrome associated with complete agenesis of the corpus callosum as well as widespread failure of commissural tracks throughout the CNS that also cause congenital mirror movements.291-294 In addition to the absence of the corpus callosum, all other commissures (anterior, posterior, and hippocampal) of the brain are absent.

Furthermore, two brothers with a 7 kb deletion that results in skipping exon 1 that encodes the membrane localization signal crucial for its function in axon guidance together with a separate individual harboring a homozygous 7bp deletion in exon 4 resulting in a truncated DCC all exhibit horizontal gaze palsy characterized by impaired conjugate horizontal eye movements, intellectual disability, and scoliosis characterized by a curved spine, while only the two brothers displayed CMM (Figure 1.13).293 The combination of all individuals reported to have DCC mutations enabled an estimated 42% CMM penetrance and a 26% ACC penetrance.293

Interestingly, males exhibit CMM twice as often as females. In contrast, five times more females

51 with a truncating DCC mutation display ACC.293 This is in accordance with findings showing

DCC expression levels to be dependent on testosterone.295 Combining all studies thus far published, a new repository for all reported disease-associated with DCC sequence variants was created using the Leiden Open Variation Databasehttps://databases.lovd.nl/shared/genes/DCC).293

In line with DCC’s role in guiding commissural axons of the spinal cord and similar to CMM;

DCC mutations in humans lead to mirroring of somatosensory stimuli.296 These individuals experience bilateral sensations evoked by unilateral somatosensory stimulation. Lastly, one particular DCC mutation, T1210A, identified through the Exome Aggregation Consortium

(ExAC) databases295 will be discussed in chapter 4. Although not predicted to cause CMM or

ACC using an in silico prediction algorithm,295 the spontaneous mutation, T1210A, identified in humans as well as T1210’s role in DCC stability will be addressed in chapter 4.

Figure 1.13: DCC mutation landscape identified in the human population

Location of all reported mutations depicted within the DCC protein structure. Square represents predicted loss-of-function; circles missense mutations. Red square mutations are associated with developmental split brain (DSBS), green represents CMM, blue ACC, orange CMM & ACC. Adapted with permission from Human Mutation. 2018 (Marsh et al., 2018). Copyright © 2018 by Human Mutation.

1.4.5.5 - DCC in neurological and psychiatric diseases In addition to its requirements for the development of CNS commissures, DCC is also involved in wiring the prefrontal cortex, the most evolutionary recent addition to the brain, during adolescent.297 Changes in DCC expression in humans have been linked to psychiatric conditions such as clinical depression and schizophrenia.297 In addition to regulating dopaminergic connections targeting the prefrontal cortex, small nucleotide polymorphism (SNP) in the DCC gene has been associated with an increased susceptibility for Parkinson’s disease.298

1.4.6 - Receptor regulation 1.4.6.1 - Regulated endocytosis/exocytosis Changing the growth cone’s response along the various intermediate targets require changes in receptors present at the plasma membrane as opposed to changing the different cues expressed along the growth cone’s journey. The receptor population present at the cell surface is a highly dynamic process regulated by several different mechanisms.186, 187 Among the several mechanisms involved, regulated exocytosis and proteolysis have been described for DCC.299-303

A significant pool of axonal DCC resides within intracellular vesicles.303, 304 It was demonstrated that netrin-induced depolarization of cortical neurons, and not spinal commissural neurons, lead to membrane insertion of DCC from these intracellular vesicles.304 This depolarization- dependent plasma membrane insertion of DCC was dependent on PKA, PKC, PI3K, and the exocytosis machinery. Communication between neurons requires exocytosis of neurotransmitter-containing synaptic vesicles. This, in turn, depends on the assembly of the soluble N-ethylmaleimide-sensitive factor attachment protein receptor (SNARE) complex formed by Syntaxin-1 (Sytx1), synaptosomal-associated protein of 25 kDa (SNAP25 – t-

SNARE), and vesicle-associated membrane protein 2 ((VAMP2)/Synaptobrevin – v-SNARE).305

Interestingly, developing neurons were shown to depend on the same molecular machinery to deliver guidance receptors to the growth cone plasma membrane. Specifically, DCC associates with Syntx1 and tetanus neurotoxininsensitive vesicle-associated membrane protein (TI-VAMP) in a netrin-1-dependent manner to mediate chemoattraction.303 Furthermore, this SNARE- mediated exocytosis of DCC is suppressed by the interaction of TRIM9 with SNAP25.303 Upon netrin-1 stimulation, TRIM9 releases SNAP25 to promote SNARE-mediated vesicle fusion to deliver DCC to the cell surface. Thus, TRIM9 is emerging as a master negative regulator of the netrin/DCC signaling pathway in the absence of netrin-1 by: 1) ubiquitinating Vasp,275 2) blocking DCC-FAK interaction and FAK activation,276 and 3) blocking SNARE-mediated exocytosis of DCC.303 In contrast, regulated endocytosis can switch a repulsive response to netrin-1 to an attraction by selectively removing UNC5 receptors from the cell surface.306 As mentioned above, repulsion occurs in response to netrin-1 when DCC and UNC5 heterodimerizes. Endocytosis is accomplished through the activation of PKC downstream of another netrin receptor, the G protein-coupled Adenosine 2B receptor (A2b). Here, activation of protein kinase C (PKC) triggers the formation of a protein complex that includes UNC5H1, protein interacting with C-kinase-1 (Pick1), and PKC that specifically removes UNC5A from the growth cone cell surface leading to a switch in the response to netrin-1 from repulsion/growth cone collapse to attraction/growth cone expansion.306 Importantly, PKC activation leads to colocalization of UNC5H1 with early endosomal markers.

1.4.6.2 - Proteolysis Receptor proteolysis has been well characterized as a means to control the kinetics and quality of signal produced by receptor-ligand interaction. It can control pathway activation, the spatial distribution and number of receptors present at the plasma membrane at any given

54 time.186, 187, 307 One of the first studies to demonstrate the role of cleavage of a guidance receptor as a means to attenuate the response to a guidance cue was ectodomain shedding of the netrin receptor DCC (deleted in colorectal cancer) by metalloproteases in response to netrin-1.302 This cleavage by metalloproteases was shown to attenuate netrin-induced axon outgrowth.

Metalloproteases are zinc-dependent enzymes of the Metzinccin superfamily.308 They include the secreted matrix metalloprotease (MMP) and the transmembranous A Disintegrin and

Metalloprotease (ADAM) families.309 The disintegrin domain mediates interaction with cell surface receptors. In contrast to the negative regulation of netrin/DCC signaling,

Kuzbanian/ADAM10 seem to play a positive role in Slit/Robo-mediated midline repulsion in

Drosophila as Kuz mutants phenocopies the loss of Slit or Robo.310 In addition, an increase in

Robo1 expression was observed in Kuz mutant commissural axons. Kuz/ADAM10 also regulates the cleavage of both Eph receptors and ephrin-A2 ligands.311 Ephrin-A2 binding to

Eph receptors induces a conformational change that exposes a recognition sequence for

ADAM10, resulting in cleavage of only ephrin-A2-bound to Eph and consequently converting an adhesion into repulsion.312 In contrast, specific cleavage of Eph-B converts an ephrin-B/Eph-B adhesion into axon retraction. Thus, cleavage of both ligand and receptor is involved in converting an attraction (i.e. adhesion) into repulsion.311

Kuz was identified in Drosophila for its role in regulating Notch signaling in the context of neurogenesis.313 Kuz-mediated shedding of Notch’s extracellular ectodomain triggers γ- secretase intracellular cleavage of Notch, thereby releasing its intracellular domain (ICD) from the membrane and allowing it to translocate into the nucleus to regulate transcription.314 This elegant mechanism mediated by γ-secretase cleavage of Notch during neurogenesis promotes heterogeneity in neuronal cell-types through lateral inhibition.315 Differentiating cells inhibit

55 their neighbors from doing the same through a salt and pepper pattern of pro-neural versus proliferative gene expression.316 Several studies have shown that following extracellular cleavage by metalloprotease, DCC gets cleaved intracellularly by presenilin-containing γ- secretase complex.299-301 Evidence suggests that the DCC-ICD might be targeted to the nucleus following γ-secretase cleavage to regulate transcription similar to the Notch-ICD.299

Specifically, it was shown that following DCC cleavage by γ-secretase, the intracellular domain was able to activate transcription of a reporter, suggesting a potential role for this cleavage event in regulating transcription.299 Furthermore, the DCC drosophila homologue, Frazzled, was shown to be a required transcription factor for commissural axon guidance following γ-secretase cleavage.317, 318 In addition, commissural neurons lacking presenilin acquire an inappropriate attraction to netrin-1 because of an accumulation DCC fragments at the plasma membrane and a corresponding insensitivity to Slit/Robo silencing.301 Thus, γ-secretase cleavage of DCC might serve two functions: 1) clear the membrane of DCC fragments still capable of mediating downstream signaling and 2) create a DCC-ICD capable of shuttling into the nucleus and regulate transcription. In addition to DCC, several ephrin ligands and Eph receptors undergo a similar ADAM10/ γ-secretase sequential cleavage.319-321 In contrast to the DCC-ICD shuttling to the nucleus, the ephrin-ICD binds to and activate Src family kinases and thus functions as a signaling molecule.320 Regulated proteolysis has also been shown to prevent premature repulsion in pre-crossing commissural axons. Before reaching the midline, axons express both

Nrp2 and Plexin-A1, two receptors for the repellant Sema3B, but Calpain-1-mediated Plexin-A1 cleavage prevents Semaphorin signaling to take place.199 At the midline, secreted-glial cell line- derived neurotrophic factor (GDNF) inhibits calpain-1 activity through the cell adhesion molecule NrCAM, thereby enabling Sema3B/Nrp2/Plexin-A1-mediated repulsion out of the

56 midline.322 Finally, it was shown that DCC gets cleaved by caspases in the absence of netrin leading the cell into apoptosis, thus highlighting the role of DCC as a dependence receptor. In chapter 3, we will show that a novel protease, calpain-1, already shown to play an important role in mediating Semaphorin/PlexinA1-mediated repulsion out of the spinal cord floor plate is also involved in regulating the netrin/DCC signaling pathway.

1.4.7 - Netrin-1-mediated cortical axon guidance In terms of cortical axon guidance, netrin-1 is implicated in the guidance of all four major axonal tracts within the forebrain. Its receptor, DCC, is largely confined to projection neurons within the forebrain.323 The first evidence that netrin plays a role in cortical outgrowth and guidance came from co-cultures of E15 rat neocortex explants with one of their intermediate target, the internal capsule.324 In this scenario, cortical axon outgrowth was stimulated and directed toward the internal capsule. The same study confirmed the role of netrin-1 in attracting cortical axons. Around the same time a different group found that netrin-1 secreted from the ganglionic eminences promoted cortical axon outgrowth and guidance of corticofugal neurons.325

Netrin-1 also plays a major role in attracting callosal axons. Following the attraction of pioneering axons toward the midline, netrin-1 secreted by midline glial structures equivalent to the spinal cord floor plate plays more of a secondary role in attenuating Slit/Robo-mediated repulsion in pre-crossing axons (Figure 1.14).326, 327 Thus, the formation of forebrain commissures such as the corpus callosum, similar to spinal cord commissures, is largely dependent on netrin-1 signaling. Corticospinal projections are among the longest axons of the vertebrate CNS (Figure 1.15). It is the major output from the motor cortex that connects the cortex to the spinal cord (Figure 1.15).328 Axon outgrowth and guidance of these axons happen almost exclusively after birth and reaches its final target around 3 weeks postnatally (Figure

1.14). Ventrally-derived netrin-1 diffuses from the ganglionic eminences to provide a

chemoattractive gradient for pioneering corticofugal axons projecting to the intermediate zone

where the majority of subcerebral projecting axons start their journey (Figure 1.15).324, 325, 328

The internal capsule which also expresses netrin-1 serves as an intermediate target for both CST

and corticothalamic axons where they separate from each other (Figure 1.14 & 1.15). CST

axons then descend posteriorly through the midbrain and hindbrain toward the spinal cord

(Figure 1.14). At the caudal-most part of the hindbrain they cross the midline dorsally, again

with the help of netrin-1 expressed at the midline (Figure 1.15).326, 328 After the initial guidance

of pioneering corticospinal axons however, the cell adhesion molecules NCAM and L1 take over

by mediating adhesion onto the already established axonal track of pioneering axons.327 This

midline crossing of CST axons is the reason why our limbs are controlled by the opposite side of

our brain. Interestingly, a recent study comparing the CST in mice with a conditional deletion of

DCC specifically in cortical neurons (Emx1-Cre;DCCflx/flx) to mice with a germline DCC

mutation (DCCkanga/-) found that midline crossing happened normally in Emx1-Cre;DCCflx/flx

mice.329 In contrast, no axons crossed the midline in DCCkanga/- mice, indicating a non-cell-

autonomous role of DCC during CST midline crossing since DCC was still expressed normally

in the CST’s trajectory in Emx1-Cre;DCCflx/flx mice.329 The mechanism by which DCC

functions to guide axons without being expressed at the growth cone cell surface is a novel idea

that merits further investigation.

Figure 1.14: Cortical axon guidance

(A) Schematic representation of the four major forebrain axonal tracks. Callosal neurons (CPN) (purple) send their axons through the corpus callosum (CC) to the contralateral side; corticospinal track neurons (CST)(blue) project to the spinal cord via the internal capsule (IC); corticothalamic neurons (CTA)(green) follow a similar initial trajectory via the internal capsule towards the thalamus while thalamocortical neurons (TCA)(red) are born in the thalamus and project to the cortex. (B) Callosal neurons born in superficial layers (II/III) send their axons to the contralateral side through the corpus callosum. Inset, several guidance molecules guide these axons along the way such as Sema3C and netrin-1 expressed at the midline. Slit2, Wnt5a, and Draxin also participate in guiding axons toward the midline. Adapted with permission from Neuroscience 2013 (Leyva-Díaz & López-Bendito, 2013). Copyright © 2013 by Neuroscience.

Figure 1.15: The long journey along the cortical spinal tract

Schematic representation indicating the major guidance decisions involved during corticospinal axon guidance from the motor cortex to the different targets within the spinal cord. Among the different molecules involved, netrin-1/DCC play important roles in guidaing axons toward the internal capsule as well as midline crossing. Adapted with permission from Neuroscience 2013 (Leyva-Díaz & López-Bendito, 2013). Copyright © 2013 by Neuroscience.

1.5 - Calpain As previously mentioned, Calpain-1 has been implicated in commissural axon guidance by cleaving the Semaphorin receptor, PlexiA1.199, 322 In chapter 3, we will demonstrate its role in regulating the netrin/DCC signaling pathway. There are 16 mammalian calpain genes with 14 encoding proteins with a cysteine protease domain (capn1-capn3, capn5-capn16) and the other two genes encoding regulatory subunits.330, 331 Several isoforms are ubiquitously expressed while others display tissue-specific expression.331, 332 Although generally thought of as intracellular proteases, some have reported active calpains in the extracellular space.330, 333 The intracellular localization of calpains is complex and highly variable, suggesting that their subcellular localization might be dynamically regulated and therefore an important way to regulate their function. Typical calpains are characterized by penta-EF-hand motifs in their C- terminal calcium binding domain while atypical calpains lack this region (Figure 1.15).330, 333

The ubiquitous calpains, calpain-1 and calpain-2, are the founding members of this family of proteases.330 Paradoxically, they are commonly referred to by their in vitro requirements for calcium such that μ-calpain (calpain-1) require micromolar calcium concentrations while m- calpain (calpain-2) require millimolar concentrations whereas physiological intracellular calcium levels never even reaches 1μM.330, 334 Only under pathological conditions such as apoptosis, axotomy, neurodegeneration, and ischemia do the calcium levels reach the in vitro requirements within the cell.333 Thus, other regulatory mechanisms capable of lowering the calcium requirement in vivo have been the subject of intense investigations. Calpain-1/2 are the predominant calpains in the CNS, are ubiquitously expressed in the brain, and are referred to as modulatory proteases owing to their limited proteolysis of substrates leading to distinct functions for the cleaved products.330, 335, 336 Their roles within the CNS are very diverse, ranging from

61 development,199, 337-341 apoptosis,342 synaptic plasticity,335, 343-345 as well as in several pathologies such as Parkinson’s disease,335, 346 Alzheimer’s disease,347 Huntington disease,348 and stroke.336

1.5.1 - Calpain regulation

Since intracellular calcium never reaches the levels required for activation in vitro, other means have been identified to compensate and lower the calcium requirement inside the cell.

Much progress has been made in delineating the mechanisms involved in calpain regulation.

Specifically, in the context of cell migration calpains play an important role in cell de-adhesion through cleavage of adhesion proteins.333 For its role in cell migration, calpain has received much attention in the context of cancer metastasis.349, 350 The ~80kDa Calpain1/2 large subunit requires heterodimerization with the small ~30kDa subunit (Figure 1.15), capns1, to form an active holoenzyme.330 Autolysis of the large subunit lowers the calcium requirement but is suggested to function more in the progression of activation rather than its initiation.333, 351, 352

Similarly, phospholipid and protein-protein binding lowers the calcium requirement.353, 354 In addition, several phosphorylation sites control calpain activity, whether activating or inhibiting.355, 356 Calpain-2 is activated by Erk1/2-mediated phosphorylation on Serine 50 even in the absence of calcium; a site that is absent in calpain-1.356 Interestingly, FAK acts as a bridge between calpain-2 and ERK, thus promoting calpain-2 activation.357 In contrast, PKA-mediated phosphorylation of calpain-2 is inhibitory.355 Conversely, in mature neurons, calpain-1 is responsible for cleaving PHLPP1β, an inhibitor of ERK1/2, therefore indirectly activating

ERK1/2.358, 359 Finally, calpain activity is inhibited by their best-known interacting partner, the only known endogenous calpain inhibitor, calpastatin.330, 333, 349 One calpastatin molecule can inhibit up to four calpain molecules.360 Calpastatin preferentially binds activated and calcium- bound calpain and is therefore thought of as an attenuation mechanism.333, 361, 362 Within the

CNS, calpain substrates include cytoskeletal proteins (e.g. spectrin, tubulin, tau, vinculin),345, 363-

368 synaptic/transmembrane proteins (glutamate receptors, calcium channels, amyloid precursor proteins), signaling proteins (e.g.PKC, CamKII, p35) and transcription regulators (e.g. p35, c-

Jun, NFκB).199, 335, 340, 343, 369-372 Specifically, cleavage of the αII-spectrin subunit, resulting in two breakdown products of ~150kDa and 145kDa, is temporally linked to calpain activation in a variety of cellular context.77, 345, 364, 366, 373-375 Thus, assessing the state of the ~150kDa and

145kDa αII-spectrin breakdown products is a powerful tool to assess calpain activation.364 No calpain consensus sequence has been found to have any significant value in predicting potential substrates.333, 376 Therefore, calpains are thought to cleave their substrates in disordered regions between structured domains.

Penta-EF-hand Protease domain C2-like domain T Capn1/2 domain

P P P P P P P P P P

P Capn1/2 Gly- Penta-EF-hand CapnS1 P Capn2 rich domain CapnS2 P Capn1

Figure 1.16: Domain organization of the classical calpains

The 80 kDa large subunits consist of a protease domain (green), a C2-like domain containing

several phosphorylation sites, and four EF hands that participate in calcium binding as well as well as binding with the small subunit. The small 30 kDa subunit contain a highly

flexible glycine rich region and four EF hands mediating interaction with the large subunits.

The unique (blue & red) and common (yellow) phosphorylation sites are indicated.

1.5.2 - Calpain in axon outgrowth and guidance An important signal that controls growth cone motility and therefore axon outgrowth and guidance is calcium.284, 377, 378 In Xenopus laevis spinal cord, local Ca2+ within filopodia regulates growth cone motility through cleavage of two adhesion proteins, talin and FAK.340, 378

Blocking calpain-mediated talin and FAK cleavage increased axon outgrowth in this system.

Several growth factors have been shown to regulate calpain activity in neurons such as EGF,

BDNF, and GDNF.322, 339, 356, 379, 380 EGF was shown to activate calpain-2 through ERK activation while the role of BDNF seems to be context-dependent. In the context of learning and memory in mature cortical neurons, BDNF stimulates calpain activity379, 380 whereas calpain activity is repressed during developmental axon branching.339 Calpain is suggested to be involved in neurite consolidation by preventing excessive branching.339 Branching factors such as BDNF, neurotrophin-3 (NT3), and netrin-1 inhibit calpain activity to promote branching.339

Finally, although GDNF does not display any guidance properties for growing axons, it was shown to play an important role in promoting commissural axon exit out of the spinal cord midline.322 In pre-crossing commissural axons, Sema3B receptors Nrp2 and Plexin-A1 are synthesized but calpain-1-mediated Plexin-A1 cleavage prevents it from being expressed at the growth cone cell surface.199 This prevents premature midline repulsion. At the midline, secreted-GDNF inhibits calpain-1 activity through the cell adhesion molecule NrCam.322 This initiates Sema3B-mediated repulsion and ensures a timely exit out of the midline. Finally, one study demonstrated the role of calpain downstream of another axon guidance molecule.

Sema3A-mediated growth cone collapse in cultured hippocampal neurons was shown to be dependent on ERK-mediated calpain activation.370 The role of calpain downstream of other guidance molecules will be interesting to investigate since many have the potential to influence calpain activity since they are expressed at the midline and throughout the axon’s journey.

Rationale and objectives

This thesis examines the molecular mechanisms involved in regulating netrin/DCC- mediated axon outgrowth. The studies presented in chapter 2 were motivated by the discovery that phosphorylation of the conserved residue 1418 of DCC is required for netrin-1-mediated signaling and directed axon outgrowth.206 In addition, the observation that the majority of DCC phosphorylation occurs on ser/thr residues led to the studies presented in chapter 4. Finally, the observation of several endogenous DCC fragments of low molecular weight led to studies presented in chapter 3. Here, the molecular mechanisms by which Y1418 phosphorylation regulates netrin/DCC signaling are explored via the characterization of a novel protein interaction with DCC. The role of a novel protease in netrin/DCC-mediated axon outgrowth as well as the role of thr 1210 phosphorylation is also explored.

In an effort to characterize novel players regulating this signaling pathway during rat cortical axon outgrowth, the specific objectives of the thesis are as follows:

Objective 1 (Chapter 2): Characterize the functional interaction between p120RasGAP and DCC downstream of netrin-1

The discovery that the N-terminus of p120RasGAP interacts with DCC upon the phosphorylation of Y1418 not only raises questions concerning the function of p120RasGAP, but also regarding the regulation and the role of Ras activity during netrin-1-dependent axon guidance. To answer these questions, the molecular mechanisms that regulate the interaction

65 between p120RasGAP and DCC are examined and the function of p120RasGAP is assessed during netrin-1-dependent axon outgrowth and chemoattraction.

Objective 2 (Chapter 3): Characterize the role of Calpain1/2 in netrin-DCC-mediated axon outgrowth

The observation presented in chapter 4 that calpastatin partially rescues the cleavage of a phospho-null DCC mutant (DCC-T1210V) as well as previous studies implicating calpain-1 in commissural axon guidance199, 322 raised the possibility that some of the endogenous DCC fragments observed on SDS-PAGE was produced by calpain-1/2 cleavage. To address this possibility, we characterized the role of calpain-1/2 in netrin/DCC-mediated axon outgrowth.

Objective 3 (Chapter 4): Characterize the role of Thr 1210 phosphorylation during cortical axon outgrowth

The identification of four ser/thr phosphorylation sites led to the discovery of Thr 1210 and the observation of a DCC fragment when replaced with an alanine residue. This led to the hypothesis that Thr 1210 phosphorylation is essential for DCC stability. By introducing other mutations at this particular site and investigating their behavior in cell culture and primary neurons we characterized the role of this novel phosphorylation site.

Preface to Chapter 2

Post-translational modifications such as phosphorylation create binding sites that mediate dynamic protein-protein interactions. The phosphorylation of the conserved Y1418 residue of DCC is essential for netrin-1 signal transduction in the central vertebrate nervous system. The rationale of the study presented in Chapter 2 is to characterize the function of proteins that interact with DCC following the netrin-1-dependent phosphorylation of Y1418 in rat embryonic cortical neurons. Mass spectrometry has proven to be a valuable tool in the search for novel protein- protein interactions and was used here to search for rat embryonic brain proteins that interact with DCC upon the phosphorylation of Y1418. Members of the Ras subfamily of GTPases are essential regulators of cellular function and integrate signals from various cellular compartments, including signaling from plasma membrane receptors. It has been generally assumed that Ras activity mediates ERK activation downstream of netrin-1 and DCC. Yet, the regulation and the role of Ras activity have not been investigated in the context of netrin-1/DCC signaling. The study represents the first report of netrin-1-dependent Ras activation and proposes a mechanism for the role of p120RasGAP during netrin-1-mediated axon outgrowth and chemoattraction that includes Ras-dependent and Ras-independent functions. An edited version of this chapter was accepted for publication in 2016 by the Journal of Biological Chemistry.

Chapter 2 – p120RasGAP Contributes as a Scaffolding Protein to Regulate Netrin-1-Mediated Axon Outgrowth and Attraction in a GAP-dependent

The N-terminal SH2 domain of p120RasGAP interacts with DCC via the phosphorylated Y1418 residue in vitro

To identify SH2-containing proteins that bind to the phosphorylated Y1418 (pY1418) residue of DCC, we screened an SH2 domain array using as bait a 15-amino acid (a.a.) synthetic

DCC peptide comprising pY1418 (DCC-pY1418). Amongst an array of 46 SH2 domains, the N- terminal SH2 (N-SH2) domain of p120RasGAP (Fig. 2.1A) bound to the DCC-pY1418 peptide as revealed by a colorimetric enzyme-linked immunosorbent assay (ELISA). The N-SH2 domain of p120RasGAP displayed a 1.24- and 1.36-fold increase in absorbance relative to the GST control with 50 and 100 nM of DCC-pY1418 peptide, respectively (Fig. 2.1B).

To validate the interaction between the N-SH2 domain of p120RasGAP and DCC- pY1418, purified GST fusion proteins of the individual N-SH2 and C-terminal SH2 (C-SH2) domains of p120RasGAP (Fig. 2.1A and C) were tested for their ability to bind to immobilized

DCC-pY1418 peptide in a dot blot assay (Fig. 2.1D). The N-SH2 domain was the sole domain capable of interacting with DCC-pY1418 and did not bind to a control spot of BSA (Fig. 2.1D).

GST control protein did not bind to the DCC peptide or to BSA (Fig. 2.1D), and tyrosine phosphorylation of DCC-pY1418 was confirmed with a phosphospecific antibody raised against

DCC-pY1418 (Fig. 2.1D). Thus, we conclude that the N-SH2 domain of p120RasGAP interacts directly with the synthetic DCC peptide via pY1418.

A. 1 1047 P N-SH2 SH3 C-SH2 PH C2 GAP

B. C. GST p120RasGAP N-SH2 1.36 m 1.4

n 2 5 2 5 2 5 μg )

0 1.3 1.24

4 a

t 1.2 35

r c e 1.1

n 25

c i

n 1 1

d a

l 1

r f

( GST N-SH2 C-SH2

o 0.9

s b

a 0.8 50 nM 100 nM DCC-pY1418

D. DCC- IB: pY1418 GST

DCC- pY1418

BSA

-- GST N-SH2 C-SH2 Figure 2.1: The N-terminal SH2 domain of p120RasGAP interacts in vitro with DCC via phosphorylated tyrosine Y1418

(A) p120RasGAP contains a proline-rich region (P), N- and C-terminal Src Homology 2 (N-SH2, C-

SH2), Src Homology 3 (SH3), pleckstrin homology (PH), calcium-dependent phospholipid-binding

(C2) and GTPase-activating protein (GAP) domains. (B) A synthetic DCC peptide containing phosphorylated Y1418 (DCC-pY1418) was used as bait to screen a SH2 domain array by ELISA. Binding of p120RasGAP N-SH2 with 50 and 100 nM of DCC-pY1418 peptide is represented as the

fold increase in absorbance relative to the absorbance obtained with a GST control. (C) 2 and 5 ug of purified GST, GST-p120RasGAP N-SH2 and C-SH2 were resolved by SDS-PAGE and the proteins

stained with Coomassie Blue. (D) The DCC-pY1418 peptide was spotted onto nitrocellulose

membranes with BSA as a control, and each membrane was incubated with either purified GST, GST-p120RasGAP N-SH2 or C-SH2 (100ng/ml) followed by immunoblotting with anti-GST antibodies. One membrane was immunoblotted with phosphospecific anti-DCC-pY1418 (DCC- 69 pY1418) antibodies (IB: immunoblot).

Netrin-1 promotes the association of p120RasGAP with DCC in embryonic cortical neurons We next examined the interaction of p120RasGAP and DCC in dissociated embryonic day 18 (E18) rat cortical neurons, which are a good cellular model to investigate netrin-1/DCC- induced signaling pathways in the context of axon outgrowth and guidance 183, 273, 323-325, 381.

DCC and p120RasGAP co-immunoprecipitated and the interaction peaked after 10 minutes of stimulation with netrin-1 (Fig. 2.2A and B). Then, we evaluated the localization of DCC and p120RasGAP by immunostaining cortical neurons following netrin-1 stimulation. Visualization by confocal microscopy revealed that p120RasGAP and DCC were both present in the cell bodies, axons, and growth cones of cortical neurons (Fig. 2.2C). Netrin-1 increased the fluorescence intensity of p120RasGAP and DCC in the axons and growth cones (Fig. 2.2C).

Quantification of the mean Pearson’s correlation coefficient (r) revealed that the correlation between p120RasGAP and DCC fluorescence intensity was significantly increased in growth cones after 10 minutes of netrin-1 stimulation (r=0.24 versus r=0.16, p=0.028), while netrin-1 treatment resulted in no significant change of the co-association (r=0.27 versus 0.24, p>0.05) in whole cells (Fig. 2.2D). Altogether, we identified p120RasGAP as a novel DCC binding partner in embryonic cortical neurons, and we demonstrate that netrin-1 promotes the recruitment of p120RasGAP to DCC preferentially in growth cones.

Figure 2.2 p120RasGAP interacts with DCC in netrin-1-induced embryonic cortical neurons

E18 embryonic rat cortical neurons were stimulated with netrin-1 (500 ng/ml) for the indicated times after being cultured 2 days in vitro (2DIV). (A) p120RasGAP was immunoprecipitated (IP) from cell lysates

with anti-p120RasGAP antibodies or mouse immunoglobulin G (IgG) as control. IP proteins and total cell lysates (TCL) were resolved by SDS-PAGE and immunoblotted (IB) with antibodies against the indicated

proteins. (B) Quantitative densitometry (±SEM) of DCC co-IP with p120RasGAP is represented as the fold change relative to 0 min of netrin-1 stimulation for at least three independent experiments. Unpaired

Student’s t test: ***p<0.005. (C) E18 embryonic rat cortical neurons (2DIV) were incubated with 500 ng/ml netrin-1 (n) or left unstimulated (-) for 10 minutes, immunostained with antibodies against DCC

and p120RasGAP, and imaged by confocal microscopy. Arrows indicate cell bodies and arrowheads indicate growth cones. The gray dashed outlines represent the growth cones. Scale bars=50 μm (wc) and

20 μm (gc). (D) The correlation between DCC and p120RasGAP fluorescence intensities in (C) was measured with the Metamorph software using the Pearson’s correlation coefficient (±SEM) in whole cells

(wc) and growth cones (gc) in three independent experiments (number of neurons = 65, 51, 54, 40 from left to right). Unpaired Student’s t test: ns, not significant; *p=0.028.

p120RasGAP associates with a DCC multiprotein signaling complex in netrin-1-stimulated cortical neurons

We next examined whether p120RasGAP is tyrosine phosphorylated in response to netrin-1. We observed that p120RasGAP was tyrosine phosphorylated in cortical neurons after 10 minutes of stimulation with netrin-1, concomitant with its association with DCC (Fig. 2.3A and B). Moreover, activated ERK (pERK) and FAK (FAK-pY397) co-immunoprecipitated with p120RasGAP and DCC in response to netrin-1 stimulation (Fig. 2.3A). Together, these results show that the assembly of a DCC- p120RasGAP protein complex with ERK and FAK is induced by netrin-1 in dissociated rat cortical neurons.

A. B. IP: RasGAP IgG TCL 10% netrin-1 (min) 0 10 0 0 10 4

i IB: t

a pY (RasGAP) r * P 3

A )

RasGAP 130 G e

s g

a n

R 2

/ a

)

DCC 250 h

P c

d l

o 1

s f

P 130

I (

FAK- a

- o

(

c pY397

Y 0 p 0 10 pERK netrin-1 (min) 35

Figure 2.3 p120RasGAP associates with a DCC multiprotein complex in netrin-1-stimulated cortical neurons

(A) Cortical neurons were stimulated with netrin-1 for 10 minutes. p120RasGAP was IP from cell

lysates with anti-p120RasGAP antibodies or mouse IgGs as control. IP proteins and total cell lysates (TCL) were resolved by SDS-PAGE and immunoblotted (IB) with antibodies against the indicated

proteins (pY (RasGAP): anti-phosphotyrosine antibodies). Data were taken from the same film exposure. The white line between the RasGAP IP and IgG lanes is indicating that irrelevant lanes

have been digitally removed from the original image. (B) Quantitative densitometry (±SEM) of pY

(RasGAP)/RasGAP ratio is represented as the fold change relative to 0 min of netrin-1 stimulation for at least three independent experiments. Unpaired Student’s t test: *p<0.05.

p120RasGAP is required to maintain basal Ras and ERK activities in cortical neurons To further characterize the function of p120RasGAP in cortical neurons, endogenous p120RasGAP expression was downregulated in E18 rat cortical neurons by electroporating synthetic siRNA targeting the 5’ end of p120RasGAP mRNA (RASA),382 which led to a significant decrease of p120RasGAP expression compared to control siRNA (Fig. 2.4A and B).

Then, we investigated the role of p120RasGAP, a negative regulator of Ras through its C- terminal GAP domain 383, in Ras activation in neurons by immunofluorescence using anti-Ras-

GTP antibodies and confocal microscopy (Fig. 2.4C). To validate the anti-Ras-GTP antibodies, we first monitor the levels of Ras-GTP in cortical neurons stimulated with nerve growth factor

(NGF), a well-established activator of Ras in neurons 384, 385. In cortical neurons stimulated with

NGF for 15 minutes, we observed a significant 2-fold (p<0.05) increase in Ras-GTP fluorescence intensity (Fig. 2.5C and D), which was comparable to the activation of Ras detected by G-LISA assay in NGF-treated cortical neuron lysates (Fig. 2.5E). In unstimulated p120RasGAP-deficient cortical neurons, Ras-GTP levels showed a significant 1.56- (p<0.005) and 2.03-fold (p<0.001) increase in whole cells and growth cones, respectively (Fig. 2.4F and

G). Next, we monitored the phosphorylation levels of ERK, a major signaling pathway activated downstream of Ras, by immunofluorescence and confocal microscopy in p120RasGAP-deficient neurons (Fig. 2.4H). We quantified the average fluorescence intensity of pERK at the plasma membrane of cortical neurons. In p120RasGAP-depleted neurons, pERK was significantly increased by 1.35- (p<0.05) and 1.53-fold (p<0.05) in whole cells and growth cones, respectively

(Fig. 2.4I). No change in total ERK fluorescence intensity was detected in p120RasGAP- deficient neurons (Fig. 2.4J and K). Overall, p120RasGAP depletion in cortical neurons caused aberrant activation of Ras and ERK in whole cells and in the neuronal growth cones. Therefore, these results demonstrate the involvement of p120RasGAP for the proper regulation of basal Ras and ERK activities in cortical neurons.

A. C. D. E. CTL RASA :siRNA IB: NGF NGF

130 y

t i

s 2.5 * 1.50

RasGAP )

a i

2.0 n * e

ezrin )

e e

70 d

s l

n 1.25

a a

e 1.5

t s

B. c

1.5 o

(

o i

r 1.0

)

l m

u 1.00

(

g -

i 0.5

T n

**** 4

R a

o 1.0 s

h 0.0 0.75

c P

R 0 15 0 15

d A

l NGF Time (min) Time (min)

( a

y 0.5

p d 0.0 CTL RASA siRNA F. G.

control siRNA RASA siRNA ) 2000 *** U

Ras-GTP A **

i s

n 1500

a n

R 1000 ▪ ▪ e

▪ c

u l

f 500 ( ) ) ) ) L A L A T S T S (C A (C A R R .c ( .c ( w .c g .c w g

H. I. )

▪ 1000 U

control siRNA RASA siRNA A

y t

pERK i *

s *

n 750

c p

n 500

u l

f 250 ( ) ) ) ) L A L A T S T S (C A (C A R R .c ( .c ( w .c g .c w g J. K. ns ) 750

control siRNA RASA siRNA U

A ,

EERRKK y

ERK t i

s ns

▪ n

▪ K

n i

E 500

s e ▪ r 21

▪ o

u l

f 250 ( ) ) ) ) L A L A T S T S (C A (C A R R ▪ .c ( .c ( w .c g .c w g ▪

Figure 2.4 p120RasGAP is required to maintain basal Ras and ERK activities in cortical neurons Control (CTL) or p120RasGAP (RASA) siRNA was electroporated in neurons at 0DIV with pGFP as a transfection reporter plasmid. (A) Total cell lysates were resolved by SDS-PAGE and immunoblotted (IB) with antibodies against p120RasGAP and ezrin as a loading control. (B) Quantitative densitometry of (A) is represented as the fold change (±SEM) relative to control siRNA measured in ten independent experiments. Unpaired Student’s t test: ****p<0.0001. (C) Cortical neurons were stimulated with NGF (100 ng/ml) or left unstimulated (-) for 15 minutes. Neurons were immunostained with anti-Ras-GTP antibodies. Scale bar=10 μm. (D) Ras-GTP fluorescence intensity and the fold increase (±SEM) relative to unstimulated control neurons was measured in at least three independent experiments. (number of neurons > 50 per condition) Unpaired Student’s t test: *p<0.05. (E) The levels of Ras-GTP in each cell lysate were evaluated by G-LISA assay by measuring the absorbance at 492 nm which is represented as the fold change (±SEM) relative to the unstimulated lysate (0 min) in at least three independent experiments. Unpaired Student’s t test: *p<0.05. (F) Neurons were immunostained with anti-Ras-GTP antibodies. Arrows indicate cell bodies and arrowheads indicate growth cones of GFP-expressing neurons. Squares represent untransfected neurons. Scale bar=50 μm. (G) Ras-GTP fluorescence intensity (arbitrary units, AU; ±SEM) of GFP-expressing neurons in (F) was measured in whole cells (w.c) and growth cones (g.c) in three independent experiments (number of neurons=31, 32, 31, 32). Two- way ANOVA, Fisher’s least significant difference (LSD) posttest: **p<0.0004, ***p<0.0001. (H) The neurons were immunostained with antibodies against pERK. Arrows indicate cell bodies and arrowheads indicate growth cones. Scale bars=50 μm. (I) pERK fluorescence intensity (arbitrary units, AU; ±SEM) of GFP- expressing neurons in (H) was measured in whole cells (w.c) and growth cones (g.c) in three independent experiments (number of neurons: 35, 36, 30, 35). Two-way ANOVA, Fisher’s LSD posttest:*p<0.05. (J) Neurons were immunostained with antibodies against ERK. Arrows and arrowheads indicate the cell bodies and growth cones of GFP-expressing neurons. Squares indicate untransfected neurons. Scale bars=50 μm. (K) Total ERK fluorescence intensity (arbitrary units, AU; ±SEM) of GFP-expressing neurons in (J) was measured in whole cells (w.c) and growth cones (g.c) (number of neurons: 41, 38, 41, 38) in three independent experiments. Two-way ANOVA, Tukey’s posttest: ns, not significant.

76 p120RasGAP is required for netrin-1-dependent attraction of embryonic cortical neurons

To determine whether p120RasGAP regulates netrin-1-dependent chemoattraction, we evaluated the impact of p120RasGAP depletion on cortical growth cone turning in response to a netrin-1 gradient using a Dunn chamber turning assay 207, 386, 387. The growth cones of cortical neurons electroporated with control siRNA randomly turned with no particular preference for any direction when exposed to a control phosphate-buffered saline (PBS) gradient (Fig. 2.5A), but were attracted to a netrin-1 gradient (Fig. 2.5B). The introduction of p120RasGAP siRNA inhibited the attractive response to netrin-1 and growth cones reverted to turning randomly (Fig.

2.5C and D). 64% of the control growth cones were attracted to the netrin-1 gradient, whereas the percentage that turned towards the netrin-1 gradient in p120RasGAP-deficient neurons (46%) was similar to the percentage that turned towards PBS in control neurons (43%) (Fig. 2.5E). In fact, the turned angle of growth cones in response to netrin-1 (9.22° ± 4.03º: mean angle turned ± standard error of the mean (SEM)) was significantly reduced when p120RasGAP was depleted (-

4.39° ± 3.86º) (Fig. 2.5F). p120RasGAP downregulation did not have an effect on displacement rates during the time the growth cones were imaged (Fig. 2.5G). These results demonstrate that p120RasGAP is required for netrin-1-dependent chemoattraction.

A. control siRNA (PBS) B. control siRNA (netrin-1) 00°° 0° --3300°° 3300°° -30° 30°

--6600°° 6600°° -60° 60° 1205 20 1015 15 5 10 10 --9900°° 5 90° -90° 5 90° repulsion attraction

C. RASA siRNA (netrin-1) D. RASA (n) 0° CTL (n) 00°° -30° 30° --3300°° 3300°°

-60° 60° --6600°° 6600°° 20 1205 15 1015 10 10 -90° 5 90° -90° 55 90° E. F. 100 2.0% 6.0% 15 * *

9.0% ) CTL (PBS)

s e

90 e CTL (n)

r 10

20.2% g

17.2% e RASA (n) d

( 5 20.2%

80 s

e g

l 0

a n

a 70

d d

21.2% 23.3% 60º to 90º e -5

n r

30º to 60º u t

u -10 ns t 60

f 0º to 30º o

n 34.8%

o -30º to 0º i

t 50 u

b -60º to -30º i G.

r 22.2%

t 25

s i 40 26.7% -90º to -60º ) ns ns

d CTL (PBS)

i e

m 20 g

/ CTL (n)

t µ n 30 16.9% ( RASA (n)

e 15

e e

p 25.3%

20 13.8% m 10

a l

p 5 s

10 16.9% i d 0 9.1% 12.9% 0 2.2% CTL (PBS) CTL (n) RASA (n)

Figure 2.5. p120RasGAP is required for netrin-1-dependent attraction

Control (CTL) or p120RasGAP (RASA) siRNA was electroporated with pGFP as a transfection reporter plasmid in E18 embryonic rat cortical neurons at 0 DIV. At 2 DIV, the neurons were exposed to a control PBS or a 200 ng/ml netrin-1 VI-V (n) gradient for 90 minutes. (A-C) Rose histograms represent the distribution of turned angles of cortical growth cones when exposed to a control PBS (A) or a netrin-1 gradient (B, C). Responses of individual neurons were clustered in 10° bins and the percentage of total neurons per bin is represented by the radius of each segment (number of neurons=99, 89, 116, respectively). (D) The overlay of the rose histograms presented in (B) and (C) is illustrated to compare the response to netrin-1 of control and p120RasGAP-deficient neurons. (E) The bar graph represents the turned angles percentage distribution of cortical growth cones in panels (A- C). (F) The mean turned angle (±SEM) toward the gradient was measured in degrees for each condition. (G) The mean displacement (±SEM) for a 90-minute netrin-1 treatment was calculated. One-way ANOVA, Fisher’s LSD posttest: ns, not significant; *p< 0.05.

The N-terminus of p120RasGAP is sufficient to mediate netrin-1-dependent cortical axon outgrowth We then explored the role of p120RasGAP in netrin-1-induced axon outgrowth in cortical neurons 183, 273. p120RasGAP siRNA was electroporated together with GFP cDNA as a reporter to visualize the neurons. We measured the average axon length of GFP-expressing neurons after

24 hours of incubation with netrin-1. Netrin-1 significantly increased the average axon length compared to unstimulated control neurons (Fig. 2.6A, top panels and B), as previously described

183, 273, 381. In unstimulated neurons, depletion of p120RasGAP resulted in significantly longer axons (103.8 μm, p<0.0001) (Fig. 2.6A, top panels and B), which correlated well with the higher levels of activated Ras and ERK detected in p120RasGAP-depleted neurons (Fig. 2.4).

However, netrin-1 decreased the length of p120RasGAP-deficient neurons (88.6 μm, p=0.0051)

(Fig. 2.6A, top panels and B). To assess the specificity of the p120RasGAP effects on axon extension, we analyzed the glutamate response in p120RasGAP-deficient neurons. Consistent with previous studies, glutamate stimulation increased axon outgrowth in cortical neurons and the effect was independent of p120RasGAP 183, 273 (Fig. 2.6C). Thus, the lack of stimulation in

79 axon outgrowth of p120RasGAP-deficient neurons was specific to netrin-1, suggesting that p120RasGAP is required for netrin-1 to positively regulate axon extension.

To determine the domains within p120RasGAP responsible for the response to netrin-1, we expressed siRNA-resistant human full-length (FL) or the N-terminus domain (NT) of p120RasGAP in control or p120RasGAP-deficient neurons (Fig. 2.6A and B, and Fig. 2.7). Re- expression of p120RasGAP-FL restored the basal axon length in p120RasGAP-deficient neurons, while it inhibited netrin-1-induced axon outgrowth in both control and p120RasGAP- depleted neurons (Fig. 2.6A and B). In contrast, re-expression of the N-terminus (NT) of p120RasGAP lacking the GAP domain was sufficient to rescue netrin-1-induced cortical axon extension (p<0.001) in p120RasGAP-depleted neurons (Fig. 2.6A and B). Thus, the N-terminal scaffolding SH2-SH3-SH2 domains mediate the positive regulation of netrin-1-dependent axon outgrowth by p120RasGAP, while overexpression of p120RasGAP blocks the response of axons to netrin-1. This is in agreement with our findings that p120RasGAP is necessary to maintain basal Ras-GTP and pERK levels (Fig. 4), which is a requisite for axon extension 382.

A. Control siRNA RASA siRNA unstimulated netrin-1 unstimulated netrin-1 GFP

vector

B. ns C. ns ns ns *** 125 (-) ***

125 (-) )

*** *** (n) ) m

*** m (g)

µ (

*** (

h ***

t t

g 100

g n

n 100

l e

*** *** l

n n

o **

o x

ns x

a a

e 75 ns e

g 75

A A 50 50 v v L L T T ; ; F F N N L A ; ; ; ; CTL RASA T S L A L A C A T S T S siRNA R C A C A R R siRNA; plasmid

Figure 2.6. The N-terminus of p120RasGAP is sufficient to mediate netrin-1-dependent cortical axon outgrowth

Control (CTL) or p120RasGAP (RASA) siRNA was electroporated in E18 embryonic rat cortical neurons at 0DIV with pGFP as a transfection reporter plasmid. Neurons at 1DIV were incubated with 200 ng/ml

netrin-1 (n), 50 μM glutamate (g) or left unstimulated (-) for 24 hours and axon outgrowth was assessed in GFP-expressing neurons. (A) Control vector (v), full-length (FL) or N-terminal (NT) p120RasGAP were

co-expressed with control or p120RasGAP siRNA and pGFP in cortical neurons. Scale bar=50 μm. (B, C) ± Axon outgrowth was measured and expressed as the average axon length (μm; SEM) in at least three independent experiments (number of neurons: in (B) 381, 224, 272, 191, 228, 180, 146, 162, 222, 165, 179,

182 and in (C) 299, 210, 381, 190). Two-way ANOVA, Fisher’s LSD posttest: ns, not significant; **p<0.005, ***p<0.001.

A. Control siRNA RASA siRNA

▪ * vector

▪

human * p120RasGAP-FL

B. ns

300 * ns

) 275 * v U

A 250 FL ,

y 225

i P

s 200

e t

G 175

n i

a 150

R c

0 125

2 e 1 100

p s

e 75 r

o 50

l f

( 25 0 82 CTL RASA siRNA

Figure 2.7. Human p120 RasGAP is expressed in p120RasGAP-deficient rat embryonic cortical neurons Control (CTL) or p120RasGAP (RASA) siRNA was electroporated in E18 embryonic rat cortical neurons at 0 DIV with control vector or full-length (FL) human p120RasGAP with pGFP as a transfection reporter plasmid. (A) The cortical neurons from Fig. 6A were immunostained with anti-p120RasGAP antibodies.

Asterisks indicate GFP-expressing neurons and squares indicate untransfected neurons. Scale bar=50 μm. (B) p120RasGAP fluorescence intensity (arbitrary units, AU; ±SEM) of GFP-expressing neurons in (A) was measured (number of neurons=38, 37, 45, 24). Two-way ANOVA, Tukey’s posttest: * p<0.0001.

The N-terminus of p120RasGAP interacts with the C-terminus tail of DCC Since the N-terminus of p120RasGAP mediates netrin-1-induced axon outgrowth, we

next investigated the molecular interaction between the SH2-SH3-SH2 domains of p120RasGAP

and DCC. GST-p120RasGAP domains were incubated with protein lysates from HEK293 cells

overexpressing either wild-type DCC or DCC mutant proteins (Fig. 2.8A and B). DCC proteins

did not bind to the GST protein control (Fig. 2.8C). p120RasGAP N-SH2 interacted with DCC,

DCC-Y1361F, and with DCC 1-1421 lacking the P3 region (Fig. 2.8A and D). However, its

binding to DCC 1-1327 truncated before the P2 region was strongly reduced (Fig. 2.8A and D).

Surprisingly, DCC-Y1418F was still able to interact with p120RasGAP N-SH2 (Fig. 2.8D).

Since FAK interacts with both the P3 region of DCC and p120RasGAP 265, 388, 389, we suspected

that DCC-Y1418F might be indirectly pulled-down by p120RasGAP N-SH2 via an interaction

with FAK (Fig. 2.8G). The phosphorylation of Y397 on FAK has been shown to mediate its

interaction with the N-SH2 domain of p120RasGAP 389. Indeed, GST-N-SH2 pulled down FAK-

pY397 along with DCC and DCC mutant proteins (Fig. 2.8D), suggesting that p120RasGAP N-

SH2 is also able to interact with DCC and FAK independently of DCC-Y1418. Furthermore, the

83 expression of either DCC 1-1421 or DCC 1-1327 severely impaired the phosphorylation of FAK on Y397 in total cell lysates, confirming that the P3 region of DCC is important for the phosphorylation of FAK on Y397 in HEK293 cells (Fig. 2.8B), as previously reported 166, 265.

Consequently, the expression of DCC 1-1421 or DCC 1-1327 impaired or completely abolished the interaction of FAK-pY397 with GST-N-SH2, respectively. It also suggested that phosphorylated Y1418 able to interact directly in vitro with p120RasGAP N-SH2 (Fig.1) may mediate the binding of DCC 1-1421 lacking the P3 region with GST-N-SH2 (Fig. 2.8D).

It has been previously shown that the 2 SH2 domains of p120RasGAP simultaneously bind to two adjacent tyrosine residues on binding partners such as p190RhoGAP and tyrosine kinase receptors 390-393. Thus, we examined whether p120RasGAP C-SH2 could also interact with a phosphotyrosine residue in the intracellular domain of DCC. We selected Y1361 as a candidate binding site for the C-SH2 since it is the closest to DCC-Y1418 (Fig. 2.8A). GST-C-

SH2 interacted with DCC, DCC-Y1418F, and DCC 1-1421 proteins, while its interaction with

DCC 1-1327 and DCC-Y1361F was impaired (Fig. 2.8E), demonstrating that p120RasGAP C-

SH2 preferentially interacted with DCC via the phosphorylated Y1361 residue. FAK-pY397 was also pulled-down indirectly with p120RasGAP C-SH2 via DCC or DCC-Y1418F interaction, but its interaction with DCC 1-1327, DCC-Y1361F, or DCC 1-1421 was severely reduced (Fig.

2.8E). Finally, the SH2-SH3-SH2 domains of p120RasGAP interacted with FAK-pY397 and

DCC, DCC-Y1418F, -Y1361F, and 1-1421, but much less with DCC 1-1327 (Fig. 2.8F).

Altogether, these data demonstrate cooperative binding of the N- and C-SH2 domains of p120RasGAP with the C-terminus region of DCC and FAK (Fig. 2.8G).

A. B. C. P P2 P3 1120 1 1445 DCC

Y1418F

F F

7 1 1

P 8

2 2 6

1120 P2 P3 1445 DCC-Y1418F 1

F F

3 4

1 C

1 7

4 3

8 1

1 1

2 2

1 1

- -

1 6

4 3

C IB:

1 D 1 Y

Y1361F Y

4 3

1 1

1 1 - - 250

P IB: DCC

Y Y 1 1 1120 P2 P3 1445 DCC-Y1361F D 1 DCC 250 130 FAK- 130 FAK- pY397

P S 1120 P2 1421 DCC 1-1421 pY397

1 u a

tubulin e

55 c P n 1120 1327 DCC 1-1327 TCL 10% o 1 P GST

D. E. F.

F F F F F F

1 7 1 1 7 7

8 1 1 8 1 8

2 2 2 2 2 2

1 6 6 1 6 1

4 3 4 4 3 3

C C C

4 3 3 4 3 4

1 1 1 1 1 1

C C C

1 1 1 1 1 1

- - - - -

- IB:

Y Y 1 1 D Y 1 Y Y 1 1 Y 1 D D IB: IB: 250 DCC 250 DCC 250 DCC 130 FAK- 130 FAK- 130 FAK-

pY397 pY397 pY397

S S

u u

a a

e e

c c

n n

o o

P P N-SH2 P C-SH2 SH2-SH3-SH2 (NT) G.

C-SH2 N-SH2

pY1361 pY1418 P P2 P3 1120 1 1445

FAK-pY397

N-SH2

Figure 2.8. The N-terminus of p120RasGAP interacts with the C-terminus of DCC

(A) The intracellular domain of rat DCC (a.a 1120-1445) contains three conserved regions (P1, P2,

P3). The conserved tyrosine residue in phosphodeficient mutants DCC-Y1418F and DCC-Y1361F was substituted for a phenylalanine residue. Truncation mutants DCC 1-1421 and 1-1327 are

truncated before the P3 or the P2, respectively. (B) DCC, DCC-Y1418F, DCC-Y1361F, DCC 1- 1421 and DCC 1-1327 were expressed in HEK293 cells. Proteins from cell lysates were pulled-

down using purified GST control protein (C) or GST-p120RasGAP N-SH2 (D), C-SH2 (E), or SH2-SH3-SH2 (NT) (F). Associated proteins and total cell lysates (TCL) were resolved by SDS-

PAGE and immunoblotted (IB) with antibodies against DCC, FAK-pY397 and tubulin. GST fusion

proteins were stained with Ponceau S. (G) The N-SH2 and C-SH2 domains of p120RasGAP interact with phosphorylated (p) Y1418 and Y1361, respectively. Alternatively, the N-SH2 can also interact with FAK-pY397 and DCC independently of DCC-Y1418. 85

Preface to Chapter 3

The netrin-1 receptor, DCC, is a substrate for several proteases such as metalloproteases, γ-secretase, caspases and others not yet identified. The increase in DCC cleavage observed upon netrin-1 stimulation prompted us to investigate the nature of this cleavage and characterize its role in the context of cortical axon outgrowth. The identification of calpain-1 as a protease cleaving DCC in response to netrin-1 initiated the studies presented in chapter 3. Calpain’s role in neurons has been well-characterized in the adult. However, few studies looked into calpain’s role during nervous system development. This is the first report demonstrating a role for calpain-1/2 downstream of netrin-1 stimulation. This is also the first report characterizing calpain-1/2’s role in mammalian cortical axon outgrowth during CNS development. The results presented in chapter 3 will be submitted for publication shortly.

Chapter 3 - A role for the calcium-activated protease calpain in the regulation of netrin-1/DCC-mediated cortical axon outgrowth and guidance

Calpain-1, Calpain-2, and Calpastatin are co-expressed with the netrin-1 receptor DCC during cortical axon outgrowth

Since the two major calpain isoforms, calpain-1 and calpain-2, have previously been shown to play an important role in cortical synaptic plasticity in mature neurons335, 343, 344 and some evidence suggests a function in axon outgrowth and branching,339, 340, 370, 378, 394 we first examined their subcellular expression along with their endogenous inhibitor, calpastatin in dissociated embryonic cortical neurons. We decided to look at early stages of neuronal development when axon specification and outgrowth occurs in dissociated cortical neurons (days in vitro (DIV) 1-10)12, 13, 66, 68 (Figure 3.1). Interestingly, calpain-1, -2, and calpastatin expression was detected along with DCC throughout neuronal development during axon outgrowth (DIV1-

3), branching (DIV3-5), and dendritic spine maturation (DIV7-10) (Figure 3.1). We found that calpain-1 was the major isoform expressed at DIV2 in the axon and to a lesser extent in the growth cone (Figure 3.2). However, its expression decreased during neuronal development with a lower expression at DIV7 (Figure 3.2B). In contrast, lower calpain-2 expression was detected at DIV2, which correlated with lower active-ERK1/2 (Figure 3.2B), although its expression increased as neurons matured in cell culture showing a higher expression at DIV7 when axon branching and dendritic spines are more developed, similar to DCC expression and active-

ERK1/2 (Figure 3.2B). In contrast to calpain-1, calpastatin expression was highly enriched within the neuronal growth cone (Figures 3.1 & 3.2). Calpastatin was expressed in the axon shaft and at the periphery of the growth cone (Figure 3.2, top panel inset) whereas calpain-1 was expressed along the axon shaft but limited to the central region of the growth cone (Figure 3.1B,

87 middle panel inset). These results suggest that calpain-1 might be active in the growth cone, though calpastatin may be limiting its activity within the peripheral region of the growth cone to promote axon outgrowth. Finally, we also found that calpain-1/2 and calpastatin were expressed in the DCC-positive cortical axonal tracks within the developing cortex at embryonic day 17.5

(Figure 3.2C). Taken together, these results show that both calpain-1 and calpain-2 isoforms, together with their endogenous inhibitor calpastatin, are expressed during cortical development, which overlaps with the expression of the netrin-1 receptor DCC.

Figure 3.1: Expression of calpain-1, calpain-2, calpastatin and the netrin receptor, DCC during cortical neuron development in vitro

Dissociated E18 rat cortical neurons were plated at DIV0 and fixed at different DIVs. Calpain-1 (top panel), DCC (top-middle panel), calpastatin (bottom-middle panel), and calpain-2 (bottom panel) expression were examined at DIV3, DIV6, DIV7, and DIV10 by immunofluoresecence. 88 Scale bar=10 μm

DIV: 2 3 7 250 DCC

100 Calpain1

Calpain2 70 Tubulin

pERK1/2

Figure 3.2: Expression of calpain-1, calpain-2, calpastatin and the netrin receptor, DCC, have

distinct but overlapping subcellular localization during cortical axon outgrowth at 2 days in vitro (DIV2).

(A) E18 rat cortical neurons were fixed at DIV2 and immunostained with the indicated antibodies and visualized with fluorescent microscopy. Inset show a closer view of the growth cones. (B) Cortical neuron lysates were lysed and loaded on SDS-PAGE for western blot analysis with the indicated antibodies. Each lane represents different stages of development in vitro (DIV3, DIV7, and DIV10). (C) E17.5 rat cortices were fixed and immunostained with the indicated antibodies on coronal sections. scale bar = 10 um; Inset scale bar=5 um; scale bar in (C)=20 um.

Netrin-1 differentially regulates the localization of calpain-1 and calpastatin in growing cortical axons

Because it was previously shown that netrin-1 stimulation affects DCC expression within the growth cone and at the cell surface;272, 303, 381 we next assessed whether netrin-1 regulates the localization of calpain-1/2 and calpastatin in dissociated cortical neurons by indirect immunofluorescence. Neurons were treated with netrin-1 over a 30-minute period, and then fixed and stained with antibodies against calpain-1, -2, and calpastatin. We observed a rapid and transient change in the expression/localization of calpain-1 and calpastatin but not calpain-2 upon netrin-1 stimulation (Figure 3.3). Calpain-1 was significantly increased along the axon shaft after 5 and 15 minutes of stimulation (Figure 3.3A, middle panel, and Figure 3.3B). On the other hand, calpastatin was significantly decreased in the growth cone after 5 minutes of netrin-1 stimulation, which was recovered to basal levels after 15 minutes (Figure 3.3A, top panel, and

Figure 3.3B). Therefore, these data demonstrate that netrin-1 differentially influences calpain-1 and calpastatin expression in extending cortical axons.

90 n

1.5 Calpastat i CalpastatiCalpastatni

Capn1 Capn2 0 ** 0

= 2 * NS 0 = 1.5

t 1 t = NS

o **

t o

t 1.5 t

o NS

e Axon

t e

g 1

g e

n Growth

n g

a 1

a n

h 0.5

h c

Cone a

c h

d 0.5 0.5

F F

0 o 0 1 2 3 4 F 0 1 2 3 4 Netrin (min): 0 5 15 30 Netrin (min): 0 5 15 30 Netrin (min): 01 52 135 340

Figure 3.3: Netrin-1 stimulation increases calpain-1, but not calpain-2, expression while decreasing calpastatin expression in axons and growth cones

(A) Dissociated E18 rat cortical neurons were stimulated at DIV2 for the indicated time.

Expression of calpain-1/2 and calpastatin were examined following netrin-1 stimulation.

Pixel intensity was analyzed and quantified in (B). scale bar = 10 um. Unpaired Student’s t

test: *p < 0.05 **p < 0.01.

Netrin-1 activates calpain in cortical neurons and stimulates the cleavage of cytoskeletal proteins in a calpain-dependent manner

Netrin-1 stimulation is known to activate focal adhesion kinase (FAK),165, 166, 206 the extracellular signal-regulated kinases (ERK1/2),279, 395 the small GTPase Rac1,183, 266, 270, 271 and the ezrin-radixin-moesin (ERM)273 proteins within 5 minutes of treatment in cortical neurons.

Because this activation of the netrin-1/DCC signaling pathway correlates with the changes in the localized expression of calpain-1 and calpastatin, we next determined whether netrin-1 regulates calpain activity in cortical neurons. To assess this, we took advantage of one of the well- characterized calpain substrate: α-spectrin, which migrates specifically as a doublet near 150 kDa upon calpain cleavage.364 The levels of this doublet can be used as a means to measure calpain activity within whole cell lysates. Cortical neurons were treated with netrin-1 for 5 minutes and we determined the levels of calpain-mediated α-spectrin cleavage by western blot. We detected a

4-fold increase in the calpain-specific α-spectrin cleavage (Figure 3.4A and B). In addition, we also observed an increase in FAK cleavage, another well-known substrate of calpain, following 5 minutes of netrin-1 stimulation (Figure 3.4A and B). Both cleavage of spectrin and FAK were abolished in the presence of calpastatin, confirming that the observed cleaved products were calpain-specific (Figure 3.4A and B). To confirm that netrin-1 stimulation activated two of its known targets, we looked at the phosphorylation status of both FAK and ERK1/2 using phosphospecific antibodies as a readout of the activated netrin/DCC signaling pathway.166, 180, 279, 381

Indeed, phosphorylation of both FAK and ERK1/2 were increased within 5 minutes of netrin-1 stimulation (Figure 3.4A). Interestingly, calpastatin treatment led to a reduction of ERK1/2 activation upon netrin-1 stimulation (Figure 3.4A). To determine the effect of netrin-1 stimulation and measure calpain activity in real time we incubated live cortical neurons with a

92 cell-permeable calpain substrate that emits a fluorescent signal only upon calpain cleavage

(Figure 3.4C).396 Within minutes of adding netrin-1 to the media we observed a significant increase in the fluorescence compared to the vehicle control (Figure 3.4D and E). Thus, netrin-1 stimulation activates calpain leading to the cleavage of cytoskeletal-associated proteins in cortical neurons concomitantly with FAK and ERK1/2 activation.

93 ii i

u i

Untreated Calpastat n

Netrin(min): 0 5 0 5 7 Untreated Calpastat n

6 0

250 = 5 *

t o

t 4

Calpain-specif c e g

130 spectrin cleavage n 3

h c

l o

F 1 130

0 Netrin: 01 52

130

2.5 Calpain-specif c Untreated Calpastat n FAK cleavage

2 **

= t

1.5

g n

55 1

d l

o 0.5 F 35 0 Netrin (min): 01 52 Boc-Leu-Met-CMAC (non-f l orescent)

GSH

Boc-Leu-Met-MAC-SG (non-f l orescent)

Calpain

Boc-Leu-Met MAC-SG (non-f l orescent) (f l orescent)

1.9 MOCK Netrin

o 1.8

t l

u 1.7 m

i 1.6

s r

e 1.5

f a

e 1.4

a e

r 1.3

n i

1.2

l o

F 1.1 Fold increase after stimulation after increase Fold

1 0 2 4 6 8 10 12 14 16 Time (min)

t t a Figure 3.4: Netrin-1 activates calpain in dissociated cortical neurons

(A) Dissociated E18 rat cortical neurons were incubated with a purified calpastatin peptide or mock treatment for 1 hour prior to netrin-1 stimulation for 5 minutes at DIV2. Western blots of whole-cell lysates are showing FAK and ERK1/2 activation as controls together with spectrin and FAK cleavage upon netrin-1 stimulation. (B) Quantification of Spectrin and FAK cleavage depicted in (A). (C) Schematic representation of the fluorogenic calpain substrate t-boc depicting the fluorescent signal generated upon calpain cleavage. (D) Representative images of live cortical neurons stimulated with netrin-1 or DMSO in the presence of the t-boc . (E) Quantification of the time-lapse stimulation showing significant increases in the t-boc signal at different timepoints. Scale bar = 10um. Unpaired Student’s t test:*p < 0.05.

Netrin-1 activates calpain in an ERK1/2-dependent manner in cortical neurons

Previous studies in non-neuronal cells have demonstrated the formation of an

ERK1/2/FAK/Calpain-2 protein complex together with the direct phosphorylation and activation of calpain-2 by ERK1/2.355, 357 Therefore, we next explored if the activation of calpain downstream of netrin-1 stimulation was dependent on ERK1/2 in cortical neurons. To assess this, cortical neurons were incubated with the ERK1/2 inhibitor U0126 prior to netrin-1 stimulation (Figure 3.5). In the presence of DMSO control, netrin-1 was able to activate calpain and ERK1/2 within 5 minutes of stimulation (Figure 3.5A and B) while netrin-1-induced calpain and ERK1/2 activities were both inhibited in the presence of the U0126 inhibitor (Figure 3.5A and B). Thus, netrin-1 activates calpain in an ERK1/2-dependent manner in developing cortical neurons.

95 i

DMSO U0126 Netrin(min): 0 5 0 5

250 Spectrin 7 U0126 6 Unst mulated Calpain-specif c 0

spectrin cleavage = t 5

o *

130 t

e 4

g n

a 3

h c

d 2

l o

F 1

Tubulin 0 55 Netrin (min): 01 52

35 pERK1/2

Figure 3.5: Netrin-1-mediated calpain activation is ERK1/2-dependent.

(A) Dissociated E18 rat cortical neurons were incubated with the ERK1/2 inhibitor, U0126, or DMSO for 1 hour prior to netrin-1 stimulation for 5 minutes at DIV2. Western blots of whole-cell lysates are

showing phosphorylated-ERK1/2(activated), αII-spectrin cleavage and tubulin as loading control. Quantification of cleaved αII-spectrin is shown in (B). Unpaired Student’s t test:*p < 0.05. N = 3

The netrin-1 receptor DCC is a novel calpain substrate in developing cortical neurons

Since netrin-1 activates calpain in cortical neurons leading to the cleavage of

cytoskeletal-associated proteins and since we observed DCC proteolysis, we next explored

whether DCC could be a substrate of calpain in response to netrin-1 stimulation. First, we

examined whether calpain-1 was able to cleave DCC in vitro (Figure 3.6A). DCC was expressed

in HEK293 cells and immunoprecipitated DCC was incubated with purified calpain-1 in the

presence of calcium. Indeed, calpain-1 was able to cleave the entire pool of DCC, which was

detected by immunoblotting using an antibody recognizing the extracellular domain of DCC

(Figure 3.6A). Since the observed proteolytic fragment was larger than the 130kDa ectodomain

96 of DCC, we concluded that the cleavage took place within the intracellular domain (ICD) of

DCC (Figure 3.6A). Then, we determined whether endogenous calpain was able to cleave overexpressed DCC in HEK293 cells. To assess this, DCC was co-transfected with a plasmid encoding calpastatin or empty vector in HEK293 cells. To identify the intracellular fragments of

DCC, we used an antibody recognizing the cytoplasmic tail of DCC (Figure 3.6B). In the absence of calpastatin, we detected five major fragments of around 55, 50, 45, 35, and 30 KDa.

Since the molecular weight of the intracellular domain of DCC is approximately 35 KDa, the proteolytic fragments of 55 to 45 KDa correspond to the cleavage of DCC at the extracellular side of the protein. In the presence of calpastatin, the presence of the fragments above 45 KDa were not affected whereas the lower fragments of 35 and 30KDa disappeared, indicating that these proteolytic fragments corresponded to calpain-dependent DCC cleavage (Figure 3.6B). On the contrary, the 35 and 30 KDa bands correspond to intracellular proteolytic fragments of DCC

(Figure 3.6B). Next, we investigated whether DCC was also cleaved by calpain in a netrin-1- dependent manner in cortical neurons. After 5 minutes of netrin-1 stimulation, we observed a net increase in the fragments of approximately 55 and 35 KDa, corresponding to the extracellular and intracellular cleavage, respectively (Figure 3.6C). Interestingly, only the netrin-1-stimulated induction of the 35 KDa fragment was abolished in the presence of calpastatin (Figure 3.6C).

These data indicated that the cytoplasmic domain of DCC was cleaved by calpain in cortical neurons, which correlated with the activation of calpain by netrin-1 at 5 minutes of stimulation

(Figure 3.6A). Taken together, we show that netrin-1-mediated calpain activation induces the cleavage of its receptor DCC while calpastatin limits the extent of this proteolysis in cortical neurons.

i i

C VecBtor DCC Calpastat i n - + - + DCC Calpain1: - +

250

DCC

Calpastat n Netrin (min): 0 5 0 5 Extracellular Intracellular

250 (low exposure) 100

35 Ant -DCC

28 (high exposure)

Figure 3.6: Netrin-1 activates calpain leading to proteolysis of its receptor DCC

(A) Immunoprecipitated DCC was incubated with purified calpain-1 in the presence of excess

calcium for 30 minutes. Proteins were then separated from agarose beads with 5X Laemmli buffer and loaded on SDS-PAGE for western bot analysis. (B) HEK293 cells were transiently

transfected with DCC and calpastatin or control plasmid for 20 hours. Whole-cell lysates were loaded on SDS-PAGE for western bot analysis. An anti-DCC targeting the C-terminal of the

intracellular domain was used to visualize intra-cellular cleavage events. Arrows indicate fragments with similar levels whereas asterix indicate rescued fragment in the presence of

calpastatin (C) Dissociated E18 rat cortical neurons were incubated with calpastatin or a mock

treatment for 1 hour prior to netrin-1 stimulation for 5 minutes at DIV2. Western blots of whole- cell lysates are showing ERK1/2 activation in in the mock treatment. Schematic representation of

the different cleavage events previously reported to occur upon netrin-1 stimulation (metalloprotease & γ-secretase) and the observed bands. Predicted antibody binding on the C-

temrinal tail is also displayed.

Calpain is a negative regulator of cortical axon outgrowth while calpastatin promotes axon extension

To investigate the role of calpain during netrin-1-induced cortical axon outgrowth, cortical neurons were electroporated with control or two siRNAs targeting calpain-1 and calpain-

2 along with a GFP reporter to identify the electroporated cells. Neurons were fixed and immunostained for endogenous calpain-1/2 two days after electroporation, showing a significant reduction of the levels of calpain-1/2 in the GFP+ versus GFP- neurons (Figure 3.7A). Similarly, a significant decrease of calpain-1 expression was observed in the calpain-1/2 siRNAs- electroporated neurons compared with control siRNA (Figure 3.7B). We then measured the

99 average axon length of GFP-expressing neurons after 24h of incubation with netrin-1. As shown previously,180, 183, 273 netrin-1 significantly increased the average axon length compared with unstimulated control neurons (Figure 3.7C and D). Depletion of calpain-1/2 in unstimulated neurons was sufficient to induce axon outgrowth to the same extent as netrin-1-stimulated neurons. However, netrin-1 was not able to further increase the axon length of calpain-1/2- deficient neurons (Figure 3.7C and D).

Ctl Capn1/2 siRNA siRNA

Calpain-1

70 Tubulin 55

1.6 *** **

1.4 * )

l 1.2

t n

g 1

e l

o 0.8

n d

o 0.6

l a

n 0.4

m e

r 0.2

n [

( 0 Netrin: - + 1 - + siRNA: ctl capn1/2

100

Figure 3.7: Calpain limits cortical axon outgrowth

(A) Control (CTL), calpain-1, or calpain-2 siRNA was electroporated in E18 embryonic rat cortical neurons at 0DIV with pGFP as a transfection reporter plasmid. Immunofluorescence was performed 48 hours later (DIV2) with the indicated antibodies to confirm the efficiency of knockdown in GFP+ cells. (B) Total cell lysates of DIV2 cortical neurons were loaded on SDS-

PAGE and analyzed by western blot to confirm an efficient knockdown of calpain-1. (C) Control (CTL) or calpain-1/2 siRNA was electroporated in E18 embryonic rat cortical neurons at

0DIV with pGFP as a transfection reporter plasmid. Neurons at 1DIV were incubated with 500 ng/ml netrin-1 or left unstimulated for 24 hours and axon outgrowth was assessed in GFP-

expressing neurons. (D) Axon outgrowth was measured and expressed as the average axon

length normalized to E.V. in at least three independent experiments. scale bar = 50 um. Unpaired Student’s t test: *p < 0.05 **p < 0.01.

Since calpastatin is an endogenous calpain inhibitor expressed throughout cortical neuron development (Figures 3.1 & 3.2), we next investigated its role in netrin-1-induced axon outgrowth. A calpastatin expression vector together with GFP cDNA were electroporated in cortical neurons left unstimulated or stimulated with netrin-1 for 24 hours. Neurons were fixed and immunostained for calpastatin, showing overexpression of calpastatin in the GFP+ neurons compared to empty vector (Figure 3.8A). Similar to calpain-1/2-depleted neurons, overexpression of calpastatin in unstimulated neurons led an increase in axon length comparable to netrin-1-stimulated neurons (Figure 3.8A and B). Moreover, netrin-1 stimulation did not promote axon outgrowth in calpastatin-expressing neurons similar to what we observed in calpain1/2-depleted neurons. Altogether, these results demonstrate that calpain activity negatively regulates axon outgrowth in cortical neurons. The lack of stimulation in axon outgrowth of calpain-1/2-deficient neurons or overexpressing calpastatin by netrin-1 further

101 supports that calpain activity is required for netrin-1 to promote axon extension in cortical neurons.

) l

o * r

t *

n 1.5 **

e t

l 1

l A

a 0.5

n ( M 0 Netrin: - + 1 - + E.V. Calpastat i n

Figure 3.8: Calpastatin promotes axon outgrowth by inhibiting calpain1/2 activity (A) Empty vector (E.V.) or calpastatin was electroporated in E18 embryonic rat cortical neurons at 0DIV with pGFP as a transfection reporter plasmid. Neurons at 1DIV were incubated with 500

ng/ml netrin-1 or left unstimulated for 24 hours and axon outgrowth was assessed in GFP- expressing neurons. (B) Axon outgrowth was measured and expressed as the average axon length normalized to E.V. in at least three independent experiments. scale bar = 50 um. Unpaired Student’s t test: *p < 0.05 **p < 0.01.

102

Preface to Chapter 4

Previous studies from our lab have demonstrated the requirement Y1418 phosphorylation for netrin-1 signal transduction in the central vertebrate nervous system. The majority of DCC phosphorylation occurs on ser/thr residues. However, the role of these phosphorylated residues have has never been explored. The rationale of the study presented in chapter 4 is to identify and characterize the function of these phosphorylated residues. When one of the identified residues was mutated to an alanine, the cleaved fragment observed on SDS-PAGE prompted us to explore this particular residue. The results presented in chapter 4 are the first to provide evidence for a role of phosphorylation in regulating DCC stability and maintain a full-length DCC.

103

Chapter 4 - Investigating the role of DCC-Thr 1210 phosphorylation during cortical axon outgrowth

Identification of four novel Ser/Thr phosphorylation sites on the cytoplasmic tail of DCC

DCC is highly phosphorylated in response to netrin-1 stimulation206 and tyrosine 1418 phosphorylation by Fyn, in particular, is crucial for netrin-mediated axon outgrowth and guidance.206 The majority of the phosphorylation, however, occurs on serine/threonine residues206 but its role in DCC function remains to be determined. Using mass spectrometry analysis, we identified 4 Ser/Thr phosphorylation sites in the cytoplasmic tail of DCC: Ser 1178,

Ser 1182, Thr 1210, and Thr 1219 (Figure 4.1). Interestingly, Ser 1178 corresponds to an

ERK1/2 consensus phosphorylation site which was later validated in a separate study.278 Since it was previously been shown that netrin-1 stimulation leads to Erk1/2 phosphorylation and activation; Erk1/2 likely phosphorylates DCC in response to netrin-1-induced activation since direct interaction between the two proteins was also previously shown.278 Surprisingly, expression of DCC-T1210A in several cell lines led to a truncated protein of 141.1 kDa compared to the full length protein of 183.2 kDa (Figure 4.1). Re-sequencing the cDNA encoding DCC-T1210A in the pRK5 plasmid confirmed that the complete coding region was intact, without any additional stop codon to explain the truncated protein. Furthermore, DCC-

T1210A was only recognized with an antibody targeting the extra-cellular domain whereas an intra-cellular targeting antibody was unable to detect DCC-T1210A. Therefore, we conclude that DCC-T1210A is cleaved within the cytoplasmic tail and hypothesized that phosphorylation of T1210 protects DCC from proteolytic cleavage.

104

A B

Figure 4.1: Identification of four novel Ser/Thr phosphorylation sites on the cytoplasmic tail of DCC (A) Using mass spectrometry analysis, we identified three novel Ser/Thr phosphorylation sites on

the cytoplasmic tail of DCC in overexpressed DCC in NIE-115 cells. The three sites were mutated with alanine for further investigation. (B) When overexpressed in HEK293 cells, DCC-T1210A mutant undergoes proteolysis, leading to a fragment of 141.4 kDa.

Phosphorylation on Thr 1210 protects DCC from proteolysis

To test this hypothesis we first replaced the alanine residue on T1210 with another

hydrophobic residue, valine. Conversely, to mimic the phosphorylated state of DCC, we

introduced a phosphomimetic residue, glutamic acid. Similar to DCC-T1210A, DCC-T1210V

105 gets cleaved but displays a different molecular weight on SDS-PAGE (Figure 4.2). In contrast,

DCC-T1210E behaves similarly to DCC-WT and does not appear to undergo any proteolytic cleavage (Figure 4.2). Thus, the negative charge provided by the phosphate on Thr 1210 is sufficient to protect DCC from proteolysis. -- DCC-T1210 --

T1210A DCC-T1210 Phospho-null E.V. DCC A V E Phospho-null Vector DCC A V E T1210V 250 T1210V 250 DCC T1210V PhosphoPhospho-null-null 130 T1210E 130 T1210E DCC T1210E PhosphoPhospho- - mimicking mimicking Blots: αDCC

Figure 4.2 The negative charge provided by the phosphorylation of T1210 is sufficient to protect DCC from proteolysis

(A) Thr 1210 was replaced by a valine to validate the cleavage observed in DCC-T1210A and a glutamic acid was introduced to mimic to negative charge provided by the phosphorylation. (B) Total cell lysates of HEK293 cells overexpressing the three DCC mutants were loaded on SDS-PAGE and

analyzed by western blot with an antibody recognizing the extracellular domain of DCC.

DCC-T1210V and DCC-T1210E are targeted to the plasma membrane

Before testing the functional relevance of this cleavage, we asked if these mutants were able to make their way to the cell surface where they can bind netrin-1. Using a cell surface biotinylation approach, we labelled all cell surface proteins with EZ-link Sulfo-NHS-LC-biotin.

This compound is cell-impermeable and labels all accessible lysine residues. After labelling

DCC-transfected HEK293 cells, we lysed the cells and pulled-down labelled proteins with avidin-conjugated agarose beads. Western blot analysis revealed that DCC-WT, DCC-T1210V, and DCC-T1210E were all being localized at the cell surface whereas DCC-T1210A was not

106

(Figure 4.3). Thus, DCC-T1210A gets cleaved before making it to the cell surface while DCC-

T1210V either makes to the cell surface after cleavage or gets cleaved at the plasma membrane.

-- DCC-T1210 -- DCC-T1210 E.V. DCC A V E Vector DCC A V E 250 250 Cell surface DCC 130 130

250 250 Lysate 130 130

Blot:Blots: αDCC αDCC

Figure 4.3: DCC-T1210V and DCC-T1210E are targeted to the plasma membrane

Transfected HEK293 cells were incubated with EZ-link-Sulfo-NHS-LC-Biotin for 30 minutes at 4oC to label all extracellular proteins. Cells were then lysed and extracellular proteins were then purified using the biotin tag and streptavidin-conjugated beads. Proteins were then separated using SDS-PAGE and DCC detected by western blotting.

Ectodomain shedding of DCC

Since DCC was previously shown to be cleaved by metalloproteases, leading to ectodomain shedding in the extra-cellular space, we asked if DCC-T1210A was being secreted in the media. When the media of DCC-transfected HEK293 cells were immunoprecipitated with an antibody recognizing the extracellular domain of DCC, we observed a fragment of DCC being released in the media in all conditions (DCC-WT, DCC-T1210A, DCC-T1210V, and DCC-

T1210E) (Figure 4.4). Interestingly, all fragments released in the media had the same molecular weight except for DCC-T1210A (Figure 4.4), suggesting that they get cleaved by a common

107 protease at the same site. However, DCC-T1210A likely gets cleaved intracellularly by a different protease at a different site than DCC-WT, DCC-T1210V, and DCC-T1210E.

-- DCC-T1210 -- DCC-T1210 E.V. DCC A V E 250 Vector DCC A V E 250

130 130

IP of media: αDCC IPBlot: of αDCC of media: αDCC Blot: αDCC

Figure 4.4: DCC ectomain shedding in the media

Immunoprecipitation of cultured media of HEK293 cells overexpressing the different DCC mutants with an antibody targeting the extracellular domain of DCC. Media was collected 20 hours after transfection and cell debris were removed by centrifugation. Following immunoprecipitation, proteins were separated from agarose beads with 5X Laemmli buffer and loaded on SDS-PAGE for western bot analysis.

DCC full-length promotes axon outgrowth in cortical neurons

Since both DCC-T1210V and DCC-T1210E were being targeted to the plasma membrane

(Figure 4.3), we decided to test the function of Thr phosphorylation in the context of netrin- mediated cortical axon outgrowth using these two mutants and DCC-WT. Following dissection of E17.5 rat embryos and electroporation of DCC constructs with GFP as a reporter at DIV0, cortical neurons were plated and stimulated with netrin-1 the following day (DIV1). Twenty four hours later (DIV2), neurons were fixed and the mean axon length was calculated by measuring the axon length of at least 50 transfected cells (GFP+) per condition. Interestingly, overexpression of both DCC-WT and DCC-T1210E stimulated axon outgrowth compared to

108 vector control. In contrast, DCC-T1210V behaved similarly to the empty vector and did not stimulate outgrowth (Figure 4.5).

Vector DCC-WT DCC-T1210V DCC-T1210E

1.5

0.5

Mean Axon Length Axon Mean (Normalized to E.V.) to (Normalized

0 1 2 3 4

Figure 4.5: DCC full length promotes cortical axon outgrowth

Empty vector (E.V.), DCC-wt or DCC-T1210 mutants were electroporated in E18 embryonic rat cortical neurons at DIV0 with pGFP as a transfection reporter plasmid. Axon outgrowth was assessed in GFP-expressing neurons at 2DIV. (B) Axon outgrowth was measured and expressed as the average axon length normalized to E.V. in two independent

experiments. scale bar = 10 um.

DCC-T1210V gets cleaved in proximity of the mutation

To identify the identity of the protease, we first determined the location of the cleavage site by performing mass spectrometry analysis of the DCC-T1210V fragment. DCC-T1210V-

109 transfected HEK293 cells were lysed and immunoprecipitated with a DCC antibody recognizing the extracellular domain. Interestingly, we found that the cleavage of DCC-T1210V was occurring around Thr 1210. We were able to conclude that cleavage of DCC-T1210V was happening between amino acids 1201 and 1222 (Figure 4.6).

1201 ksashsgqdteeagssmstler 1222

T1210V DCC T1210V

Approx. 27 kDa

Figure 4.6: Cleavage site of DCC-T1210V mutant Immunoprecipitated DCC-T1210V was sent for mass spectrometry analysis to determine the cleavage site. DCC-T1210V gets cleaved between amino acids 1201 and 1222.

Calpastatin partially rescues DCC-1210V cleavage

We then decided to investigate the identity of the protease cleaving DCC-T1210V by incubating DCC-transfected HEK293 cells with various protease inhibitors (Figure 4.7).

110

Interestingly, when the endogenous calpain inhibitor, calpastatin, was added to the cells we observed a partial rescue of DCC-T1210V cleavage (Figure 4.7). Since we have shown in chapter 3 of this thesis that DCC is a calpain substrate (chapter 3), we hypothesised that calpain cleaves DCC and that phosphorylating Thr 1210 protects DCC from calpain-mediated cleavage.

---- Empty Vector ------DCC-WT ----

prk5 DCC

Untreated

Leupeptin

PMSF

ALLN

Leupeptin

ALLN

MG132

Untreated

MG132

PMSF

Aprotin

Calpastatin

Aprotinin Calpastatin

PMSF

Calpastatin

MG132

ALLN

PMSF

Aprotinin

untreated

Leupeptin

Calpastatin

MG132 ALLN

Aprotinin

untreated Leupeptin 250 250

130 130

100

---- T1210V ------T1210E ----

Untreated

Aprotinin Calpastatin

ALLN

Leupeptin PMSF

Untreated ALLN

MG132

PMSF MG132

Leupeptin Aprotinin Calpastatin

prk5 250

130 Blot:αDCCBlot: DCC

Figure 4.7: Calpastatin partially rescues cleavage of DCC-T1210V

HEK 293 cells were transfected with the indicated plasmid for 20 hours and incubated for 3 hours with various protease inhibitors before cell lysis. Total cell lysates were then loaded on

SDS-PAGE and analyzed by western blot.

111

Chapter 5 - Discussion 5.1 Major findings The results presented in this thesis were undertaken with the aim of identifying novel mechanisms regulating the netrin/DCC signaling pathway in the context of cortical axon outgrowth. This led to the identification of: 1) a novel DCC binding partner mediating crosstalk between distinct signaling pathways, 2) a novel protease regulating DCC, and 3) a novel phosphorylation site regulating DCC stability. The functional relevance was investigated in the context of rat cortical axon outgrowth.

5.2 p120RasGAP mediating Ras and netrin/DCC signaling

In the second chapter, we reported a novel interaction between p120RasGAP and DCC in the context of cortical neuron development in vitro. We demonstrate that p120RasGAP is essential for netrin-1-mediated cortical axon outgrowth and guidance. In addition, we show that the N-terminus of p120RasGAP positively regulate netrin-1-induced axon outgrowth by serving as a scaffolding protein mediating protein-protein interaction required for downstream signaling, independent of its GAP domain (Figure 5.1). To our knowledge, this is the first study implicating a positive role for p120RasGAP during axon outgrowth and guidance. In particular, p120RasGAP was considered an inhibitor of axon outgrowth and guidance due to its activity of its C-terminal RasGAP domain.382, 397, 398

Previous studies have implicated Ras GTPases downstream of ephrins, semaphorins, and neurotrophins during neuronal development but their role in netrin/DCC signaling had never been explored.104, 399 In particular, p120RasGAP regulates the level and duration of PDGF- dependent Ras/ERK activation in fibroblasts.400, 401 In contrast, the absence of p120RasGAP

112 recruitment to the EphB2 receptor delayed and reduced ephrin-B1-dependent ERK inhibition in neuroblastoma cells.398 Similarly, p120RasGAP plays an important role in keeping Ras/ERK inactive until netrin-1 stimulation induces p120RasGAP tyrosine phosphorylation and sequestration, therefore relieving Ras/ERK inhibition.

It was previously shown that netrin-1/DCC activate ERK whereas the canonical MAPK pathway implicates activation of ERK1/2 downstream of Ras activation.104, 279 How ERK1/2 gets activated downstream of netrin-1/DCC remained unclear. Here we show that under basal conditions, p120RasGAP puts a break on Ras, and consequently ERK activation via its GAP domain. Upon netrin-1 stimulation, p120RasGAP gets tyrosine phosphorylated along with DCC and other signaling proteins such as FAK. Interaction with DCC is achieved via p120RasGAP’s

N-terminus domain-containing SH2 domains and DCC’s phosphorylated Y1361 and Y1418 residues. In addition, previous studies have shown that phosphorylated FAK-Y397 mediates interaction with both the N-SH2 domain of p120RasGAP and DCC.265, 389 This sequesters p120RasGAP to DCC and within the multiprotein complex required for downstream signaling. p120RasGAP-bound to DCC can no longer stimulate Ras GTPase activity, which remains active, leading to ERK activation.

5.2.1 p120RasGAP’s GAP-independent function

Our findings strongly support the requirement for a tight control of p120RasGAP expression and consequently Ras activity during cortical axon outgrowth. In p120RasGAP- depleted cortical neurons, basal axon outgrowth is increased and unresponsive to netrin-1-

113 stimulated axon outgrowth. Re-expression of p120RasGAP in these p120RasGAP-depleted neurons decreases axon length to unstimulated control neurons. However, these neurons remain unresponsive to netrin-1 stimulation when p120RasGAP is overexpressed. Netrin-1-mediated axon outgrowth is only restored when the N-terminal SH2-SH3-SH2 domains of p120RasGAP is expressed in p120RasGAP-depleted neurons. This suggests that when overexpressed, p120RasGAP’s GAP domain takes over and netrin-1 stimulation cannot sequester enough p120RasGAP to prevent the GAP domain’s inhibition of Ras activity. Since p120RasGAP play two very distinct functions, a tight control on its expression and localization is likely necessary to respond adequately to extracellular cues in a timely manner. Thus, the N-terminus of p120RasGAP is necessary and sufficient to mediate netrin-1/DCC-induced axon outgrowth and guidance. This is in line with a previous study that proposed a model in which p120RasGAP is sequestered and inhibited by the neogenin receptor via an interaction with the phosphorylated

Y397 residue of FAK.382 The repulsive guidance cue RGMa releases p120RasGAP from that interaction to stimulate Ras’ GTPase activity thereby inhibiting downstream signaling. Other studies have shown that the GAP activity of p120RasGAP is inhibited when its N-terminus interacts with p190RhoGAP, FAK, p200RhoGAP, or SOCS-3.389, 402-404 Here we show that axon outgrowth and attraction mediated by netrin-1/DCC involves the reverse mechanism whereby netrin-1 stimulation favors p120RasGAP sequestration within the multiprotein complex as opposed to releasing it upon ligand stimulation. p120RasGAP seem to have evolved two very different and seemingly opposite functions: 1) putting a break on axon outgrowth via Ras inhibition via its GAP domain under basal conditions and 2) mediating netrin-1/DCC-induced axon outgrowth upon netrin-1 stimulation via its N-terminus domain.

114

Other groups have also reported GAP-independent functions for p120RasGAP.

Expression of the N-terminus increases Ras and ERK activation in fibroblasts398 and promotes

Ras-dependent differentiation of PC12 cells.405, 406 In addition, cell migration is severely impaired in p120RasGAP-deficient mouse fibroblasts, which can be rescued by the N-terminus of p120RasGAP, similarly to the rescue in cortical axon outgrowth by the N-terminus in chapter

2. This suggests that p120RasGAP is required for cell motility, both in the context of cell migration and axon outgrowth and guidance. Our results presented in chapter 2 also support a positive role for the N-terminus of p120RasGAP in Ras/ERK activation. We show that

Ras/ERK activation mediated by the N-terminus of p120RasGAP is required downstream of netrin-1/DCC signaling during growth cone navigation.

5.2.2 p120RasGAP and other GAPs

Although netrin-1 is well-known for promoting axon outgrowth;206, 251, 279, 325 it reduced axon outgrowth in p120RasGAP-depleted cortical neurons. The mechanism by which netrin-1 reduces axon outgrowth in these p120RasGAP-depleted neurons is unclear. Since p120RasGAP interacts with three different RhoGAPs, it is possible that the observed decrease in axon outgrowth upon netrin-1 stimulation is due to a loss in RhoA regulation. First, p190RhoGAP was shown to be enriched in growth cones and that the p190RhoGAP-p120RasGAP-FAK could be necessary for netrin-1-mediated axon outgrowth.407, 408 Second, interaction of p120RasGAP with another RhoGAP, deleted in liver cancer (DLC1),409, 410 which is expressed during and required for embryonic brain development,411 could also be involved in netrin-1-mediated axon outgrowth. In contrast, interaction of DLC1 with p120RasGAP inhibits DLC1’s GAP activity, thereby promoting RhoA activation, and possibly reducing axon outgrowth.412 Third, p120RasGAP interacts with the brain-specific p200RhoGAP.403, 413 It was proposed that FAK,

115 p190RhoGAP, and p200RhoGAP interact with p120RasGAP to prevent Ras inhibition by p120RasGAP.389, 403 Taken together, this suggests that Ras and RhoA regulation are interdependent and by removing a regulator of one, it likely affects regulation of the other.

Therefore, interplay between Ras and RhoA regulation is plausible in the context of netrin-

1/DCC signaling and axon outgrowth and guidance.

o t

p Netrin-1

r C

C D

ERK Ras pERK Ras GDP GTP

GAP N

NT -Y1361 -pY1361 T P p120RasGAP -Y1418 -pY1418 GA [Type a quote from the document or

pY the summary of an interesting point.

K A

A You can position the text box

F F

-Y397 -pY397 N anywhere in the document. Use the

T P Drawing Tools tab to change the GA formatting of the pull quote text box.] pY axon outgrowth growth cone attraction

Figure 5.1. Proposed model for the role of p120RasGAP in netrin-1/DCC-mediated axon outgrowth and guidance

In the absence of ligand, the GAP domain of p120RasGAP maintains Ras inactive (Ras-GDP). In response to netrin-1 stimulation, DCC-pY1418/Y1361, FAK (pY397), pERK and tyrosine phosphorylated p120RasGAP associates in a multiprotein signaling complex. Through these interactions mediated by the N-terminal SH2-SH3-SH2 domains (NT) of p120RasGAP, Ras-GTP is released to induce sustained ERK activation in response to netrin-1, leading to axon outgrowth and growth cone attraction. 116

5.3 Calpain-1/2 in netrin-1/DCC signaling

In chapter 3, we show that calpain is a novel protease regulating netrin-1/DCC-mediated signaling pathways during axon outgrowth. We demonstrate that netrin-1 regulates calpain-1 expression and its endogenous inhibitor, calpastatin, in developing cortical neurons. The increased calpain-1 expression in the axons together with the decrease in calpastatin expression correlates with the positive regulation of calpain activity induced by netrin-1 in developing neurons. Moreover, we determined that netrin-1-mediated calpain activation is ERK1/2- dependent. Of interest, netrin-1-mediated calpain activation leads to the cleavage of its receptor

DCC and cytoskeletal-associated proteins. Thus, we propose that calpain activation downstream of netrin-1/DCC-induced signaling pathways is an important mechanism to control axon outgrowth in developing neurons. In support of this, we find that calpain-1/2 is a negative regulator of cortical axon outgrowth independently of netrin-1. Both calpain-1/2 knockdown and overexpression of its endogenous inhibitor calpastatin lead to increased cortical axon outgrowth, which render the axons unresponsive to netrin-1 treatment. To our knowledge, this study is the first one to implicate calpain as a novel regulator of the chemoattractant cues netrin-1 and DCC during axon outgrowth (Figure 5.2). In the absence of netrin-1, calpain activity is a limiting factor in promoting axon extension. We propose a model in which netrin-1 activates ERK1/2, which in turn activates calpain, leading to proteolysis of its receptor DCC and cytoskeletal- associated proteins such as spectrin and FAK. The calpain-mediated intracellular fragment of

DCC may be targeted either for degradation or to the nucleus to transcriptionally regulate proteins modulating axon outgrowth (Figure 5.2).

117

4 1

N N

F Netrin F F VI EGF1 EGF2 EGF3 NTR

Netrin

N N

VI EGF1 EGF2 EGF3 N

F F

netrin NF TR

6 6

6 Netrin

N N

F netrin

F F MP

P1 spectrin P1 P1 spectrin

P2 P2 P2 Calpain Calpastat n Calpain pERK

P3 P3 P3

K A F pERK Calpastat n

Figure 5.2: Proposed model for the role of calpain in netrin-1/DCC-mediated axon outgrowth and guidance

In the absence of netrin-1, calpain activity is mostly repressed by calpastatin. Upon netrin-1 stimulation, DCC homodimerizes via its P3 domain. Extracellular metalloprotease cleaves and releases the ectodomain of DCC which primes DCC for γ-secretase cleavage. In parallel, netrin- 1 activates calpain via ERK1/2 activation which can then cleave several of its cytoskeletal targets such as spectrin and FAK, as well as DCC.

5.3.1 Netrin-1/DCC/Calpain in cortical axon outgrowth

In earlier studies calpain activity was also reported to reduce axon extension340, 378, 394 and

essential for neurite consolidation.339 In particular, it was shown that calpain activity prevented

axon branching through repression of the protrusive activity while branching factors such as

BDNF, NT3, and Netrin-1 inhibited this repression by inhibiting calpain activity and thus

allowing axonal branch formation.339 On the other hand, our findings show that netrin-1

activates calpain in cortical neurons plated after two days in culture when axon specification and

outgrowth takes place, preceding the formation of axon branches. This suggests that calpain

118 activity may be differently regulated by various extracellular factors at various stages of neuronal developmental. Netrin-1 is known for its branch-promoting activity155, 414 and for promoting axon outgrowth.183, 251, 279 It is important to note that when cortical neurons are stimulated at

DIV1 for 24 hours, netrin-1 promotes axon outgrowth.183, 206, 272, 279, 381 On the other hand, when neurons are stimulated for 72 hours, netrin-1 stimulates axon branching.155, 414 Thus, it is plausible that netrin-1 activates calpain early to prevent aberrant branching and allow the cytoskeletal machinery to focus on axon outgrowth while later in development netrin-1 inhibits calpain to promote axon branching. Looking at different developmental stages and stimulating neurons for different length of time will help clarify this. Furthermore, the activation of calpain that we observed following netrin-1 stimulation was fast and brief. Thus, since calpain activation was not sustained and netrin-1 promotes significant axon outgrowth on a different timescale (≥24 hours), we conclude that calpain activation is brief to prevent pathological responses associated with sustained calpain activation while allowing a quick way to attenuate the pathway. Finally, PKA was shown to directly phosphorylate and inhibit calpain-2 activity355 as well as being involved in the netrin-1/DCC signaling pathway by promoting DCC plasma membrane insertion in response to netrin-1.274, 415, 416 Thus, removal of DCC from the cell surface through proteolysis in response to netrin-1 leads to PKA-dependent DCC membrane insertion from intracellular pool. In parallel, inhibiting calpain-2 activity in a PKA-dependent would promote axon outgrowth.

5.3.2 Netrin-1/DCC/Calpain in cortical axon guidance

The netrin-1/DCC signaling pathway is essential for the formation of commissures within the central nervous system, regulating the axonal projections of commissural neurons in the developing spinal cord and forebrain commissures such as the corpus callosum.417, 418 Mice

119 lacking the expression of DCC have severe defects in the formation of brain commissures419 that are very similar to the phenotype of mice lacking netrin-1.420 Even a small truncation in the intracellular domain of DCC (DCCkanga)290 or deletion of its downstream effector Trio leads to severe defects in axonal projections290 with a complete absence of the anterior commissure in

Trio-depleted mice.183 Highlighting the importance of the formation of brain commissures in the human population, mutations in the DCC and NETRIN-1 genes are causative of a rare developmental disorder caused by defects in commissure formation.291, 293, 421-423 Congenital mirror movements (CMM) is characterized by early-onset, involuntary movements of one side of the body that mirror intentional movements on the opposite side in individuals with no other clinical signs or symptoms. Our findings suggest that the removal of calpain-1 and calpain-2 in the central nervous system will affect the axonal projections of the brain commissures. In support of this, biallelic mutations in CAPN1 (the calpain-1 gene) causes autosomal recessive hereditary spastic paraplegia (HSP), spastic ataxia, or both.341, 424 HSPs are a heterogenous group of inherited neurological disorders characterized by pyramidal tract affection and variable other neurological manifestations. Furthermore, locomotor defects due to abnormal neuronal development were reported in C. elegans, Drosophila, and zebrafish424 while calpain-1 null mice develop ataxia due to abnormal cerebellar development.341 Using morpholino targeting the capn1a gene in zebrafish, one group showed strong axon guidance defects in capn1a-deficient animals.424 Severe defects in motor neuron projections were observed as well as several migration defects. Specifically, in capn1a-knockdown animals, a reduction in acetylated-tubulin was observed at the level of the optic tectum and cerebellum, whereas a stronger staining was found in the dorsal telencephalon, suggesting an increase in axon outgrowth and/or branching.

This is in accordance with calpain-1 limiting axon outgrowth and branching in mice cortical

120 neurons observed in vitro by our group and others339, 340, 378, 394 It will be interesting to look into the different commissures in mice lacking the CAPN1 gene or in patients harboring a CAPN1 mutation to assess if there are any overlapping defects between DCC and CAPN1 mutations.

Alternatively, since calpain-2 might rescue the loss of calpain-1, targeting the small subunit necessary for both calpain-1 and calpain-2 activity will eliminate this problem. In fact, mice lacking the small subunit, capns1 (also known as calpain-4), have altered dendrite morphology leading to defects in synaptic transmission, learning, and memory.343 Using this previously published mouse line in which the small regulatory subunit is deleted specifically in neural progenitors, Nestin-Cre; CAPNS1 flox/flox (CAPNS1-Nestin-cKO),343 we looked at the development of cortical projections (Figure 5.3). When compared to Nestin-Cre control, there was a strong increase in TAG(+) projections within the cortex at postnatal day 3. This suggests an increase in axon outgrowth and/or branching which similar to what we and others have observed in vitro.339 In support of capain-1’s role in midline crossing, we observed a reduction in TAG+ projections within the corpus callosum, suggesting a callosal projection neurons axon guidance defect. Future studies in more animals will be needed to confirm these results. Since calpain-1 has also been implicated in commissural midline crossing in the spinal cord in the context of semaphorin signaling199, 322 and netrin-1 is well-known for its role in commissural axon attraction to the midline,194, 291, 326 it will be interesting to test if calpain-1 can mediate a crosstalk between the two signaling pathways. Using primary cultures of CAPNS1-Nestin-cKO mice, we could test if and how these signaling pathways are affected by looking at the activation state of downstream effectors. Furthermore, it will be interesting to look at different CNS projections such as the anterior commissure and commissures of the spinal cord in these

CAPNS1-Nestin-cKO as well as our preliminary in vivo results in CAPNS1-Nestin-cKO (Figure

121

5.3). Taken together, the combination of the results presented in chapter 3, together with results found by others strongly supports a crucial role for calpain-1 in normal neural circuitry assembly.

WT (Cre; +/+) midline KO (Cre; flx/flx) midline

Cortex Corpus callosum Corpus callosum

Figure 5.3: Axonal projections in CAPNS1-Nestin-cKO brain

Immunofluorescence on coronal sections of postnatal day 3 mouse forebrain with the transient axonal glycoprotein-1 (TAG-1) expressed along axons. The different structures ad regions are identified in red. Wilt-type (Nestin-Cre; CAPS1 wt/wt) is on the left whereas Nestin-Cre;CAPNS flox/flox (CAPNS1-Nestin-cKO) is on the right. Scale bar 20μm.

5.3.3 Netrin-1/DCC/Calpain and ERK activation

Similarly to our findings with netrin-1, growth factors such as EGF, BDNF, and the repulsive guidance cue semaphorin3A have been reported to activate calpain356, 379, 380 in an

122

ERK1/2 activity-dependent manner.356, 380.370. In contrast, we also show that netrin-1-induced

ERK1/2 activation is reduced in the presence of the calpain inhibitor calpastatin, suggesting that calpain activity may regulate ERK1/2 activation by netrin-1 in cortical neurons. This potential positive feedback loop is supported by another study. Using a different calpain inhibitor, calpeptin, one group have implicated calpain activity as a requirement for ERK1/2 activation under pathological conditions.375 Importantly, in mature neurons, calpain-1 is responsible for cleaving PHLPP1β, an inhibitor of ERK1/2, resulting in the indirect activation of ERK1/2.341

Therefore, calpastatin may prevent calpain-1 from relieving ERK1/2 inhibition by PHLPP1β, thereby reducing overall ERK1/2 activation upon netrin-1 stimulation. Interestingly, the defects in cerebellar development observed in calpain-1 knockout mice were rescued by crossing these calpain-1 KO with PHLPP1β KO mice.341 Thus, it will be interesting to see if removing

PHLPP1β could rescue the increased axon outgrowth observed in calpain-1/2 knockdown and calpastatin overexpressing cortical neurons. Preliminary results using cortical primary culture of CAPNS1-Nestin-cKO E15.5 embryos, support the requirement of calpain activity for ERK1/2 activation (Figure 5.4). The mechanism by which calpain is activated in an ERK-dependent manner while ERK itself requires calpain activity could very well be mediated by calpain-1- mediated cleavage of PHLPP1β, thereby activating ERK1/2. One distinction between the inhibition of calpain activity using our calpastatin inhibitor that strongly decreased netrin- dependent ERK1/2 activation and the preliminary results showing an absence of ERK1/2 activation in CAPNS1-Nestin-cKO cortical neurons is the duration of calpain inhibition. In the first scenario, calpastatin inhibits calpain transiently for one hour (+5 min netrin-1 stimulation) whereas in CAPNS1-Nestin-cKO neurons, calpain activity is constitutively repressed due to the absence of the small regulatory subunit from the neuron’s time of birth. Therefore, several

123 defects within the signaling network could affect ERK activation in these neurons, thereby preventing netrin-1-mediated ERK1/2 activation. Stimulating these neurons with other factors known to activate ERK will distinguish between a netrin-1-specific signaling defect or a general

ERK inhibition in the absence of calpain activity. Again, repeating these experiments is necessary to make any substantive conclusions. Finally, since both Rac1 and calpain activity are required for cytoskeletal re-arrangements necessary for growth cone motility and Netrin-1- mediated ERK1/2 activation has previously been shown to be at least partially dependent on

Rac1 activity;269 manipulating Rac1 and looking at calpain activity will help determine if Rac1 plays a role in regulating calpain activity.

Figure 5.4: Netrin-1-mediated ERK1/2 activation is abolished in CAPNS1-Nestin-cKO cortical neurons

Dissociated Nestin-Cre; CAPS1 wt/wt (WT) or CAPNS1-Nestin-cKO (cKO) E16 mouse cortical neurons were stimulated with netrin-1 for 5 minutes at DIV2. Western blots of whole- cell lysates are showing calpain-mediated spectrin cleavage and ERK1/2 activation in WT upon netrin-1 stimulation.

Among the two ubiquitous calpain isoforms, only calpain-2 possesses the S50 ERK phosphorylation site. However, we show that calpain-1 is the predominant calpain expressed at

DIV2, when we stimulated cortical neurons with netrin-1 and observed calpain activation. If the calpain activation we see upon netrin-1 stimulation is ERK1/2-dependent, then calpain-2 has to

124 play an important role in cleaving spectrin upon netrin-1 stimulation or calpain-1 also gets activated by ERK1/2. Alternatively, since it has previously been shown that calpain1/2 can be activated by autolysis,332, 334, 351, 352, 425 it is possible that calpain-2 gets activated by ERK1/2 downstream of netrin-1 which can then activate calpain-1. Calpain-1 can then take over by activating other calpain-1 molecules. However, since we see calpain activation within 5 minutes of netrin-1 stimulation and calpain-2 levels are much lower than calpain-1, it is unlikely that autolysis between calpain-2 and calpain-1 play any significant role. Other means of activating calpain downstream of netrin-1 stimulation is likely involved.

5.3.4 Netrin-1-stimulated calcium influx

Although intracellular calcium never reaches the in vitro levels required to maximally activate calpain, calcium influx can nevertheless play a role in potentiating the calpain activation in vivo. Since calpain-1 requires micromolar calcium concentrations whereas calpain-2 requires millimolar concentration; calpain-1 is more likely to get activated by intracellular calcium influx either from the extracellular space or release from intracellular pools. In support of this, filopodial calcium influx has been shown to influence growth cone behavior through calpain activation in Xenopus laevis.378, 426 Calcium influx was shown to decrease the extent of axon outgrowth through calpain activation in Xenopus spinal neurons. Furthermore, the influx of calcium through TRPC1-containing calcium channels located on growth cone filopodia activates calpain, which then cleaves the integrin adaptor protein talin to reduce Src-dependent axon outgrowth.394 The reduction in axon outgrowth, due to reduced adhesion upon calcium-mediated calpain activation is achieved, at least partly, by calpain-mediated FAK and talin cleavage.340

125

Several studies have implicated Src family kinase activation downstream of netrin-1 in the context of cortical axon outgrowth and guidance.165, 166, 206 Similarly, we showed calpain- mediated FAK cleavage downstream of netrin-1 stimulation. Importantly, netrin-1-mediated growth cone turning has been reported to be dependent on depolarization-dependent Ca2+ influx through TRPC calcium channels as well as Ca2+-induced Ca2+ release from cytoplasmic stores.283, 427-430 Specifically, blocking Ca2+ influx converts netrin-1-mediated growth cone attraction to repulsion.430 This Ca2+ influx and growth cone decision in terms of attraction or repulsion was shown to depend on the ratio of the small cyclic nucleotide cAMP and cGMP such that high cAMP/cGMP ratios lead to calcium influx and growth cone attraction.428, 429 Therefore, netrin-1-mediated calpain activation likely involves parallel mechanisms: 1) ERK-dependent calpain activation and 2) Ca2+ influx through TRCP1-contaning calcium channels. In addition, autolysis might further activate other calpain molecules. Thus, local calcium influx and consequently calpain activation, could promote local and asymmetrical adhesion disassembly which would lead to growth cone turning, similarly to Xenopus spinal neurons. Taken together, although not sufficient to fully activate calpain in vivo, netrin-1-induced calcium influx could participate in calpain activation via calcium influx in parallel to ERK-dependent calpain activation and allow the proper growth cone response to netrin-1.

5.3.5 DCC-ICD and transcription regulation

Following extracellular metalloprotease-mediated DCC ectodomain shedding, leaving a small transmembrane and intracellular domain (ICD) at the plasma membrane (DCC stub) with signaling capabilities.300, 301 The DCC stub is then subject to presenilin 1(PS1)-mediated

126 processing at the plasma membrane.299-301 PS1 is an essential component of the γ-secretase complex that cleaves amyloid-β precursor protein (APP), leading to the formation of toxic plaques that contribute to Alzheimer’s disease (AD).431 Inhibiting PS1 with a potent γ-secretase inhibitor or culturing primary cortical neurons from PS1-deficient mice leads to increased axon outgrowth due increased DCC signaling.300, 301 Interestingly, following γ-secretase cleavage, the

DCC-ICD was shown translocate to the nucleus and activate transcription of a reporter, suggesting the DCC-ICD possesses an intrinsic transcriptional activation domain.299 In

Drosophila, the conserved P3 domain of frazzled/DCC was shown to encode a transcription activation domain (Figure 5.5).318 This P3-activation domain activates transcription of the endosomal protein Commissureless (Comm), which reduces sensitivity to the midline repellant

Slit.317, 318 Thus, at least in Drosophila, upon γ-secretase cleavage the frazzled/DCC-ICD shuttles to the nucleus to activate transcription of comm and consequently promotes midline crossing independent of netrin-1.318 How Frazzled/DCC-ICD regulate transcription is unclear since it does not contain an obvious DNA-binding domain.318 Therefore, Neuhaus-Follini and

Bashaw (2015), hypothesized that the Fra ICD’s DNA-binding activity and specificity arise from associations with DNA-binding partners, as in the case of Notch.318

Rat DCC 1421 d lseqmasleg lmkqlnaitg saf 1445 Drosophila DCC (Frazzled) 1350 e lnqemanleg lmkdlsaita nefec 1375

Figure 5.5: Drosophila frazzled and rat DCC P3 domain alignment

The P3 domain of rat DCC shares 58% (14/24)homology to frazzled’s P3 domain in amino acids, which encodes an intrinsic transcription activation domain.

127

In mammalian cells the role of DCC-ICD is not as clear. To my knowledge, the only group that was able to visualize the highly labile DCC-ICD in mammalian cell lysates was by pre-treating HEK293cells with the proteasome inhibitor lactacystin.299 In chapter 3, we showed an increase in the production of the DCC-ICD in rat primary cortical neurons without any pretreatment, suggesting that this DCC-ICD might be stable enough to regulate transcription (Figure

5.6). Future studies will be needed to determine if this DCC-ICD fragment produced upon netrin-1 stimulation can shuttle to the nucleus and regulate transcription (Figure 5.6). Since no comm gene has been identified in mammals, other target genes will be regulated by this mammalian DCC-ICD. Although basal levels of DCC-ICD could be detected in cortical neurons, which could be mediated by γ-secretase cleavage, the increase in DCC-ICD production upon netrin-1 stimulation that we showed is mediated by calpain activity since calpastatin abolished this increase (Figure 5.6). Therefore, we show that the production of this particular

DCC-ICD detectable in cortical neurons is mediated by the novel protease, calpain, targeting

DCC upon netrin-1 stimulation (summarized in Figure 5.6).

128 i

spectrin

P1 P1

Calpastat n Calpain

P2 P2

P3 P3 3

ICD P

3 2

ICD P

P P Transcript onal regulat on?

Degradat on? nucleus

Figure 5.5: Cleavage events downstream of netrin-1 stimulation Upon netrin-1 binding to DCC, extracellular metalloprotease (MP) releases the netrin- bound ectodomain. On the intracellular side, γ-secretase cleaves DCC at the plasma membrane. In addition, activated calpain cleaves DCC, releasing the DCC-ICD and potentially shuttling to the nucleus to mediate transcription.

Several calpain-targeted therapeutic strategies have already been developed and

tremendous efforts are currently put in improving the specificity and bioavailability of calpain

inhibitors to treat neurodegenerative disorders, brain ischemia, cardiovascular disorders,

myopathies, ophthalmic diseases, cancer, infectious diseases, and recently identified

calpainopathies.350 Thus, a better understanding of the molecular mechanisms involved in

calpain regulation will contribute at improving the therapies targeting these very diverse

pathologies.

129

5.4. Phosphorylation of DCC Thr 1210 regulates its stability

In chapter 4 we identified a novel threonine phosphorylation site on the cytoplasmic tail of DCC. Our lab has showed that tyrosine 1418 phosphorylation in response to netrin-1 is required for downstream signaling and normal axon outgrowth and guidance.206 Since the majority of DCC phosphorylation occurs on Ser/Thr residues, we decided to explore their function. When we mutated one of the identified sites (T1210) to an alanine we were intrigued by the lower molecular weight observed on SDS-PAGE. Furthermore, replacing the threonine

1210 with another hydrophobic residue, valine, seemed to prime DCC for proteolytic cleavage.

By replacing the threonine with a glutamic acid we confirmed that the negative charge was sufficient to confer DCC the stability required to remain in its full-length form. Surprisingly, replacing Thr 1210 with an alanine or a valine did not result in DCC cleavage at the same location since the two mutants (DCC-T1210A and –T1210V) migrated differently on SDS-

PAGE, indicating different molecular weights. With the aim of characterizing the function of this particular phosphorylation in the context of netrin-1-mediated axon outgrowth, we verified that the mutants were making their way to the plasma membrane where netrin-1-DCC interaction takes place and the signaling cascade is initiated. Since we couldn’t detect any DCC-T1210A at the cell surface, we focused our efforts on the two mutants (DCC-T1210V and -T1210E) that were being localized at the plasma membrane for functional analyses. In addition, immunoprecipitation of the culture media of transfected cells confirmed the ectodomain shedding of DCC into the media. The DCC fragment being released into the media, along with the two mutants targeted to the cell surface were identical in size, suggesting a consensus cleavage site for ectodomain shedding. In rat embryonic cortical neurons, DCC-WT or DCC-

T1210E, but not DCC-T1210V, overexpression was enough to increase axon outgrowth in the

130 absence of netrin-1 stimulation. We determined that cleavage of DCC-T1210V was occurring between amino acids 1201 and 1222, in the vicinity of Thr 1210. Finally, preliminary results suggests that the cleavage of DCC-T1210V might be mediated by calpain since incubation of cells expressing this mutant with calpastatin prior to cell lysis partially rescued DCC-T1210V cleavage.

5.4.1 Phosphorylation as a protective mechanism

Protein phosphorylation is one of the most common and important post-translational modifications (PTMs). This reversible PTM changes the polarity of the protein giving the modified protein hydrophilic properties and allowing it to change conformation and interact with different binding partners. Most cellular processes implicate protein phosphorylation such as in protein synthesis, cell growth, cell division, and signal transduction.432 More than two-thirds of the 21,000 proteins encoded in the human genome has been shown to be phosphorylated and the majority of phosphorylation occurring on Ser/Thr. The majority of phosphorylation occurs on serine residues (86%) followed by threonine (12%) and tyrosine phosphorylation representing only 2%.433 Similarly, Ser/Thr phosphorylation represents the majority of DCC phosphorylated residues.206 DCC-T1210 is likely constitutively phosphorylated since removing the negative charge provided by the phosphate group renders DCC highly susceptible to proteolytic cleavage.

Since phosphorylation can eliminate or introduce protein binding sites, it has the potential of regulating protein-protein interactions. These interactions can either protect the phosphorylated protein from proteases or can prime the modified protein for proteolysis.434-445 Specifically, upon ligand binding, Erk1/2 phosphorylates the progesterone receptor, which targets the receptor for degradation.445 Similarly, when Cubitus interruptus (Ci), a downstream effector of hedgehog signaling, gets phosphorylated by PKA, GSK3 and casein kinase I (CSKI), Ci gets targeted for

131 proteolysis which converts the transcription activator into a repressor.439 In contrast, phosphorylation of neurofilaments, p35, and ezrin, among others, have been shown to be protective from calpain-mediated proteolysis.434, 436, 441, 444 First, dephosphorylation of the

200kDa neurofilament increased its proteolysis by calpain.441, 444 Second, a developmental switch in the phosphorylation status of p35 modifies its susceptibility to calpain-mediated cleavage.434 Embryonic p35 is highly phosphorylated in the developing brain, which protects it from calpain proteolysis and prevents p25 production. In the adult brain, unphosphorylated p35 predominates which is readily cleaved by calpain into p25. Accumulation of this calpain- mediated cleavage of p35 into p25 has been observed in cultured neurons undergoing cell death.446 Third, PKA-mediated phosphorylation on ser66 protects ezrin from calpain-1-mediated cleavage. Thus, Ser/Thr phosphorylation as a means to protect proteins from proteolysis could be a conserved mechanism regulating protein stability. Since the functional relevance of most ser/thr phosphorylation has yet to be characterized, future studies will be needed to determine if regulating protein stability through phosphorylation is a widespread mechanism inside the cell.

5.4.2 Phosphorylation modifies protein conformation

The DCC cleavage observed when Thr 1210 is replaced with a hydrophobic residue is likely due to a conformational change exposing a cleavage site normally hidden by the phosphorylated conformation. Phosphorylation can trigger transitions between conformations with different activity or binding specificity leading to distinct physiological consequences.432,

433, 447 Since we observed two distinct fragments depending on the nature of the mutation

(alanine or valine), different changes in conformation induced by the alanine and valine mutation

132 might expose different protease sites. Although preliminary results point to calpain as a candidate protease cleaving DCC-T1210V, we did not incubate cells expressing DCC-T1210V with an exhaustive number of inhibitors to conclusively determine the identity of the protease, as in Figure 4.7. In addition, we will need to repeat the experiment in Figure 4.7 to draw any solid conclusions on the rescue shown with calpastatin incubation. DCC-T1210A, on the other hand, is cleaved closer to the extracellular domain since we see a smaller fragment on SDS-PAGE as well as secreted in the cultured media when we label the fragment with an antibody targeting the extra-cellular domain. Therefore, γ-secretase might be cleaving DCC-T1210A, releasing the ectodomain and transmembrane domain into the media. Incubating cells expressing DCC-

T1210A with a γ-secretase inhibitor will help clarify if this is the case. Alternatively, changes in conformation and 3D structure of the protein upon threonine 1210 mutation might render the protein unstable and cleaved within the secretory pathway which might explain why we cannot detect DCC –T1210A at the plasma membrane.448

5.4.3 Identity of the kinase phosphorylating Thr 1210

According to NetPhos 3.1’s prediction algorithm, the likely candiate for the kinase responsible for phosphorylating T1210 is casein kinase II (CskII) (score: 0.518) and to a lesser extent GSK3 (score: 0.440). If this is the case, we predict that inhibiting CskII will render DCC unstable and susceptible to proteolytic cleavage, similarly to DCC-T1210A and DCC-T1210V.

133

5.4.4 The cytoplasmic tail of DCC is required for downstream signaling and axon outgrowth

The cytoplasmic tail of DCC is essential for downstream signaling and netrin-1/DCC- mediated axon outgrowth and guidance.206, 266, 271, 449 The increased axon outgrowth measured when DCC-WT or DCC-T1210E, but not DCC-T1210V, was overexpressed in cortical neurons suggests that overexpression artificially mimicked netrin-1 stimulation by promoting DCC dimerization and consequently downstream signaling. The DCC-T1210V cleavage of the cytoplasmic tail indicates that without this signaling platform, DCC cannot induce the cytoskeletal changes required to potentiate axon outgrowth.

5.5 Conclusion

Cell signaling cascades involve various signaling mechanisms. In this thesis, we characterize three distinct means of regulating the netrin-1/DCC signaling pathway: 1) protein- protein interaction, 2) receptor proteolysis, and 3) receptor phosphorylation. The identification and characterization of RasGAP in the netrin-1/DCC signaling pathway highlights the crosstalk between Ras and Netrin-1 signaling. The GAP-independent functions identified in this thesis add an extra layer of complexity for the role of GTPase-activating proteins during cell signaling and the netrin-1/DCC signaling pathway in particular. The identification of Calpain-1/2 as a novel protease regulating netrin-1/DCC signaling and cortical axon outgrowth highlights the importance of signaling hubs in regulating crosstalk between pathways during axon outgrowth and guidance.186, 187, 201 By implicating calpain-1 downstream of netrin-1/DCC signaling in addition to calpain-1’s already documented role in semaphoring/Plexin-A1 signaling, we provide evidence to suggests that calpain-1 might be one of these signaling hubs regulating the expression of various guidance receptors. Playing a negative role in axon outgrowth as well as

134 being involved in cell death,336, 342, 446, 450, 451 calpain could be targeted as a means to regenerate the nervous system, in addition to potential therapies in cardiovascular disorders, ophthalmic diseases, and cancer.350 Since clinical trials targeting calpain are already under way, extending our knowledge of calpain function in health and disease will help minimize side effects due to calpain inhibition. Finally, by characterizing the role of a novel phosphorylation site that was found to be sporadically mutated in the human population,295 we extend our knowledge of DCC regulation. This will help us have a better understanding of the causes of CMM and ACC in humans. Lastly, molecules targeting Netrin-1-DCC interaction are currently underway in the context of cancer452 and a better understanding of DCC regulation will help improve therapeutic strategies.

135

Materials and methods

Plasmids The plasmids pRK5, pRK5-DCC, pRK5-DCC-Y1418F, pRK5-DCC-Y1361F, pRK5-DCC 1-

1327 and pRK5-DCC 1-1421 were previously described.206, 271, 288 pRK5-DCC-P1336A/P1339A was generated by site-directed mutagenesis (Stratagene). The pCDNA3-GAP and pCDNA3-

GAP-N (human) constructs were provided by T. Pawson.401 The plasmids encoding GST and

GST- human p120RasGAP (N-SH2, SH3, C-SH2 and SH2-SH3-SH2) were provided by L.

Larose (McGill University, Montreal). pmaxGFP was purchased from Lonza (cat # VSPI-1003).

The amino acid sequence alignment of DCC orthologs was generated with the BioEdit software

Clustal W plugin. The following siRNAs were purchased: Silencer Negative Control No.1 siRNA (Life Technologies Ambion) and previously described 5’-

GCAGGGAAATCTGGAAGCTACCTTA-3’p120RasGAP siRNA (Dharmacon).382 Calpain-1

(5’-UACCUCUGUUCAAUUGCUCUA-3’) and Calpain-2 (5’-GCGGUCAGAUACCUUCA

UCAA-3’) SiRNA were purchased from Bio Basic as previously described. pRK5-DCC-

T1210A was generated by Mutagenex. The point mutations pRK5-DCC-T1210V and pRK5-

DCC-T1210E were generated with “QuikChange II Site-Directed Mutagenesis Kit” (cat#

200523). For DCC-T1210V, the forward primer: 5’CCATTCAGGTCAAGATGTCGAGGA

GGCAGGGAGCTCC-3’ and the reverse primer 5’-GGAGCTCCCTGCCTCCTCGACATC

TTGACCTGAATGG-3’. For DCC-T1210E, forward primer: 5’CCATTCAGGTCAAGATGAG

GAGGAGGCAGGGAGCTCC-3’ and reverse: 5’-GGAGCTCCCTGCCTCCTCCTCATC

TTGACCTGAATGG-3’ The calpastatin expression plasmid, pCMV-Sport6-Calpastatin, was kindly provided by Dr. Alyson Fournier (Montreal Neurological Institute, Montreal).

136

Cell culture and transfection

Cell culture was maintained in a humidified incubator at 37ºC with 5% CO2. HEK293 and COS-

7 cells were cultured in Dulbecco's modified Eagle's medium (DMEM, Wisent Bioproducts) supplemented with 10% fetal bovine serum and antibiotics. Cells were transfected overnight with pRK5, pRK5-DCC, pRK5-DCC-T1210A, pRK5-DCC-T1210V, pRK5-DCC-T1210E, pCMV-

Sport6-Calpastatin using polyethylenimine (PEI, PolyScience, cat # 23966-1) prepared according to the manufacturer’s instructions.273, 381 Briefly, HEK293 cells were plated in 100 mm-dishes and transfected when they reached 70-80% of confluency. cDNA constructs (4-8 μg) were incubated with PEI (40-80 μg) in 1 mL of DMEM for 15 minutes. The transfection mixes were then added to cells with 5 mL of fresh supplemented DMEM. After five hours transfection media was replaced with fresh media. Cells were harvested 16-20 hours post-transfection.

Cortical neurons from E18 rat embryos (Charles River) were dissociated mechanically and plated on dishes treated with poly-D-lysine (0.1 mg/ml; Sigma-Aldrich) or glass coverslips treated with poly-L-lysine (0.1 mg/ml; Sigma-Aldrich). Neurons were cultured in 10% FBS DMEM for five hours and the media was replaced with Neurobasal-A medium supplemented with 2% B27 and

1% L-glutamine (Invitrogen). Neurons were treated with the following reagents: purified recombinant netrin-1(500ng/ml), Calpastatin (calbiochem), U0126 (Cell Signaling) DMSO or water.

Antibodies

The following antibodies were used for immunoblotting and immunofluorescence: mouse monoclonal anti-DCC, clone G97-449 (BD Biosciences Inc); mouse monoclonal (AF5) anti-

137

DCC (Calbiochem); anti-GST, anti-RasGAP B4F8 and anti-DCC A-20 (Santa Cruz

Biotechnology); anti-phosphotyrosine (pY) 4G10, anti-tubulin (Millipore); anti-Calpain-1

(abcam); anti-Calpain-2 (Cell Signaling); anti-Calpastatin (Santa Cruz); anti-Spectrin, clone AA6

(milipore); anti-FAK (BS Biosciences); anti-phospho-p44/42 MAPK (Erk1/2) (Thr202/Tyr204) and anti-p44/42 MAPK (Erk1/2) (Cell Signaling Technology); anti-FAK [pY861] and anti-FAK

[pY397] (Life Technologies Novex); anti-active Ras (NewEast Biosciences); anti-DCC

[pY1418] (polyclonal antibodies raised in rabbit against the peptide KPTEDPASVpYEQDDL

(DCC-pY1418)); anti-mouse Alexa 488, anti-mouse Cy3 and anti-rabbit Cy3, anti-rabbit Alexa

555 (Life Technologies Molecular Probes); anti-goat Immunoglobulin G (IgG) Cy3 (Sigma). The following reagents were used: recombinant chick netrin-1 and netrin-1 VI-V were produced and purified as previously described 251, 453, glutamate kindly provided by D. Bowie (McGill

University), nerve growth factor (NGF) (Cedarlane).

Affinity purification and mass spectrometry

An affinity column was prepared using a phosphopeptide corresponding to amino acids 1409 to

1423 of rat DCC (Small Scale Peptide Synthesis, W.M. Keck Facility, Yale University, USA), phosphorylated on tyrosine 1418 (KPTEDPASVpYEQDDL) coupled to Affigel (Bio-Rad) according to the manufacturer’s protocol. An unphosphorylated peptide and Affigel beads were used as negative controls. Protein lysates from E13 rat brains were loaded on each column and proteins bound to the affinity columns were eluted using a gradient of sodium chloride by Fast

Protein Liquid Chromatography. The eluted proteins were separated by SDS-polyacrylamide gel electrophoresis (PAGE) and proteins were detected using the Coomassie blue gel staining method. An 85 kDa band that was only present in the phosphopeptide affinity purification was

138 cut and sent to be identified by tandem mass-spectrometry. Protein identification was made with the Mascot software (Matrix Science).

Identification of the cleavage site DCC-T1210V

Overexpressed-DCC-T1210V in HEK293 cells was immunoprecipitated and loaded on SDS-

PAGE for coomassie blue gel staining. The cleaved DCC-T1210V band was sent to the

“Proteomics Research Discovery Platform” of the IRCM for tandem mass-spectrometry.

Purification of GST fusion proteins Production of GST and GST-p120RasGAP proteins was induced with 0.5 mM isopropylthiogalactopyranoside (IPTG) for 2 h at 37°C. Bacteria pellets were resuspended in

BME-Phosphate buffer (1:1000 beta-mercaptoethanol (BME), 1 mM phenylmethylsulfonyl fluoride (PMSF), 1mM ethylenediaminetetraacetic acid (EDTA) in PBS), supplemented with 0.5 mg/ml lysozyme and incubated for 30 min at 4 °C. Resuspended pellets were frozen in an ethanol/dry ice bath, thawed in warm water, sonicated and incubated with 1% Triton X-100 for

10 min at 4 °C. Protein lysates obtained after centrifugation were incubated with glutathione- agarose beads (Sigma) for 2 h at 4°C. The beads were washed three times in BME-Phosphate buffer and stored at 4°C or proteins were eluted with 5 mM glutathione buffer in 50 mM Tris-

HCl pH 8.0 and concentrated with a Nanosep 10K Omega column (PALL) and stored at 4°C.

Purity and concentration were determined by Coomassie Blue-stained SDS-PAGE.

Peptides, SH2 domain array and dot blot assays The following peptides were purchased from Keck MS & Proteomics Resource (USA): biotinylated TEDPASVpYEQDDLSE (DCC-pY1418). The SH2 Domain-Based RTK Profiling

139

Kit (Signosis) was used according to the manufacturer’s instructions with some modifications.

Streptavidin-HRP was directly added to samples of 50 or 100 nM of DCC-pY1418 without any primary antibodies. Absorbance was read spectrophotometrically at 450 nm. Absorbance fold increase was calculated by normalizing the absorbance of each condition with the GST control absorbance. For the dot blots, 20 μg DCC-Y1418 and 300 μg BSA were spotted onto nitrocellulose membranes. Membranes were incubated with freshly purified GST or GST- p120RasGAP proteins overnight (O/N) at 4°C. The membranes were immunoblotted with anti-

GST, anti-DCC-pY1418 antibodies.

GST pull-down

Transfected HEK293 cells were lysed in 1% Triton X-100 lysis buffer as previously described

183. Protein lysates (1mg) were pre-cleared with 30 μl of glutathione–agarose beads (Sigma

Aldrich) for 2 h at 4°C and incubated with 10 or 20 μg of fresh GST or GST-p120RasGAP proteins coupled to glutathione–agarose beads for 3 h at 4°C. Beads were washed three times in ice-cold lysis buffer and boiled in SDS sample buffer.

Immunoprecipitation

Cortical neurons (2 days in vitro (DIV)) lysates were prepared as previously described 273.

Protein lysates (1mg) were incubated with 4ug of anti-p120RasGAP with protein G Sepharose beads (GE Healthcare) for 3 h at 4°C.

140

Immunoblotting and quantitative densitometry

Proteins were resolved by SDS–PAGE and transferred onto nitrocellulose membrane.

Membranes were stained with Ponceau S (Sigma-Aldrich), immunoblotted with the indicated antibodies, and visualized using enhanced chemiluminescence (Millipore) or Clarity western

ECL substrate (Bio-Rad). Membranes were visualized with ChemiDoc Touch Imaging System.

Optical density was measured using the Quantity One software (Bio-Rad). The following optical density ratios were calculated: co-immunoprecipitated DCC and p120RasGAP over immunoprecipitated p120RasGAP; pY (p120RasGAP) over p120RasGAP; p120RasGAP over ezrin; cleaved spectrin over tubulin; cleaved FAK over tubulin; pERK over tubulin;. Optical density fold change was calculated by normalizing the ratio of each condition with the control ratio.

Ras G-LISA assay

Cortical neuron (2DIV) lysates were prepared and processed as per the manufacturer’s instructions (Cytoskeleton). Absorbance was read spectrophotometrically at 492 nm. Optical density fold increase was calculated by normalizing each condition with the control’s optical density.

Immunofluorescence

Cortical neurons (2DIV) were fixed 30 min with 3.7% formaldehyde in 20% sucrose PBS at

37°C, quenched 5 min in 0.1 M glycine at room temperature, permeabilized 5 min in 0.25%

Triton X-100, and blocked 30 min with 3% BSA. Primary and secondary antibodies were incubated in 0.3% BSA. A 15 min fixation with 10% trichloroacetic acid in water was used for

141 phosphospecific antibodies 454. Neurons were examined with the following: Olympus IX81 motorized inverted microscope (40× U PLAN Fluorite and 60× U PLAN S-APO oil objective lenses) with a CoolSnap 4K camera (Photometrics); Zeiss LSM780 confocal microscope

(63×/1.40 oil Plan-Apochromat) with 488 nm argon and 561 nm DPSS lasers and a GaAsP detector. Pearson's correlation coefficient and fluorescence intensity quantification were measured with the Metamorph software.

Immunohistochemistry

E17.5 embryonic brains were fixed in 4% PFA overnight, cryopreserved for 2 days in 30% sucrose, and cryosectioned (10μm/section). Slides were then dried at 37oC before washing with

PBS. Antigen retrieval for Calpain-1 and Calpastatin was necessary for optimal staining: Slides were incubated with 10mM Sodium Citrate buffer, pH 6.0 at 90oC for 20 minutes, then cooled at room temperature and washed several times with PBS. Blocking for one hour with 5% donkey

(in PBS+0.1%Triton-X 100) serum at room temperature was performed before incubation with anti-DCC, clone G97-449 (BD Biosciences Inc), anti-Calpain-1 (abcam); anti-Calpain-2 (Cell

Signaling); anti-Calpastatin (Santa Cruz) in PBS+0.1%Triton-X 100 overnight at 4oC.

Axon outgrowth and Dunn chamber assays Axon length of GFP-expressing cortical neurons (2DIV) was measured with Metamorph.

Cortical neurons (2DIV) were plated on coverslips used for Dunn chamber assembly as previously described 387. Gradients were generated with purified netrin-1 VI-V (200 ng/ml) or buffer containing PBS in the outer well. Cell images were acquired every 3–4 min for at least 90 min on a temperature controlled stage. Neurites of at least 10 μm in length were tracked in GFP- expressing neurons. The final position of the growth cone was used to determine the angle turned

142 over 90 min relative to the gradient position. Measurements are presented in rose histograms in bins of 10° with the length of each segment representing the frequency of measurements in percent. Percentage distribution of turned angles, average turned angle and average displacement are also represented.

Time-lapse fluorescence imaging

Live cell imaging was performed with a Zeiss inverted microscope LSM780 using a 40x oil objective and 10x digital zoom. Images were acquired every 60 seconds and played every

25msecs. To maintain cell viability, the microscope was equipped with a chamber to regulate temperature and CO2. We also limited the time the cells were incubated in the microcope chamber to 45-50 minutes. For the addition of netrin-1, 50l of solution containing the amount of netrin-1 for a final dose of 500nM was applied.

Cell surface biotinylation assay

Transfected HEK293 cells were washed three times with cold PBS (+ 1mM MgCl2 +0.1mM

o CaCl2) and incubated at 4 C for 20 minutes with 5 ml of EZ-Link-Sulfo-NHS-LC-Biotin

(0.5mgml) in PBS. Cells were then washed four times with PBS (10mM glycine) at 4oC before lysis with RIPA buffer. After centrifugation, supernatants were loaded on SDS-PAGE for western blot analysis.

Calpain Assay

DCC-transfected HEK293 cells were lysed with HEPES-based RIPA buffer containing 2mM

CaCl2 without protease inhibitors. Immunoprecipitated DCC was then incubated with 1.6 ug of

143

Calpain-1 (purified from Porcine Erythrocytes - cat# 208712) for 30 minutes at 37oC. Reaction was stopped by adding 50 ul of 5X laemmli buffer.

Calpain activity assay

20 µM of BOC-LM-CMAC (Molecular Probes) was added to living cortical neurons 20 minutes prior to visualization of Calpain activity. Exposure times were kept constant and at a medium grey scale saturation to detect both increase and decrease in fluorescence. When imaged at 20 minutes after addition of tBoc, the observed intensity was highly reproducible between experiments and therefore the data were combined. After some initial trials, we determined that measurements from neuronal somas and proximal axon, as opposed to distal neurites and growth cones, would be more easily measurable. A compartment of less variable size (or width), as in the case of the soma, is more desirable than distal neurites because it minimizes the differences in fluorescence due to changes in neurite thickness. In addition, the highly dynamic nature of growth cones and distal axons make it difficult to follow and measure accurately over an extended period of time. In contrast, within the 15 minutes imaging period, the cell soma and proximal axon are relatively static and allow for more accurate measurements. After subtracting background levels, we referred every data point of the "control" and netrin-1 coverslips to a fold change to pre-treatment activity.

Statistical analysis

Statistical analysis was performed with GraphPad Prism 6 and Microsoft Excel. The data are presented as the mean ± the standard error of the mean (SEM).

144

References

1. Pandya, S.K. Understanding brain, mind and soul: contributions from neurology and neurosurgery. Mens sana monographs 9, 129-149 (2011).

2. Plato & Jowett, B. The republic, Edn. Collector's. (Easton Press, Norwalk, CT; 1980).

3. Hatfield, G.C. Perception and cognition : essays in the philosophy of psychology. (Clarendon Press ;

Oxford University Press, Oxford

Oxford ; New York; 2009).

4. Frankish, K. Quining diet qualia. Consciousness and cognition 21, 667-676 (2012).

5. Merleau-Ponty, M. & Landes, D.A. 1 online resource. (Routledge, Abingdon [England] ; New York; 2012).

6. Eagleman, D.M. Visual illusions and neurobiology. Nature reviews. Neuroscience 2, 920- 926 (2001).

7. Ramón y Cajal, S. Histologie du système nerveux de l'homme & des vertébrés, Edn. Éd. française rev. & mise à jour. (Consejo Superior de Investigaciones Científicas, Instituto Ramón y Cajal, Madrid,; 1953).

8. Kolb, B. & Whishaw, I.Q. An introduction to brain and behavior, Edn. 2nd. (Worth Publishers, New York; 2006).

9. Jessell, T.M., Kandel, E.R. & Schwartz, J.H. Principles of neural science, Edn. 4th. (McGraw-Hill, Health Professions Division, New York; 2000).

10. Glickstein, M. Golgi and Cajal: The neuron doctrine and the 100th anniversary of the 1906 Nobel Prize. Current biology : CB 16, R147-151 (2006).

11. Molnar, Z. & Brown, R.E. Insights into the life and work of Sir Charles Sherrington. Nature reviews. Neuroscience 11, 429-436 (2010).

12. Barnes, A.P. & Polleux, F. Establishment of axon-dendrite polarity in developing neurons. Annual review of neuroscience 32, 347-381 (2009).

13. Takano, T., Xu, C., Funahashi, Y., Namba, T. & Kaibuchi, K. Neuronal polarization. Development 142, 2088-2093 (2015).

145

14. Spruston, N. Pyramidal neurons: dendritic structure and synaptic integration. Nature reviews. Neuroscience 9, 206-221 (2008).

15. Lowery, L.A. & Van Vactor, D. The trip of the tip: understanding the growth cone machinery. Nature reviews. Molecular cell biology 10, 332-343 (2009).

16. Herculano-Houzel, S. The glia/neuron ratio: how it varies uniformly across brain structures and species and what that means for brain physiology and evolution. Glia 62, 1377-1391 (2014).

17. von Bartheld, C.S., Bahney, J. & Herculano-Houzel, S. The search for true numbers of neurons and glial cells in the human brain: A review of 150 years of cell counting. The Journal of comparative neurology 524, 3865-3895 (2016).

18. Harris, J., Tomassy, G.S. & Arlotta, P. Building blocks of the cerebral cortex: from development to the dish. Wiley interdisciplinary reviews. Developmental biology 4, 529- 544 (2015).

19. Haydon, P.G. GLIA: listening and talking to the synapse. Nature reviews. Neuroscience 2, 185-193 (2001).

20. Custo Greig, L.F., Woodworth, M.B., Galazo, M.J., Padmanabhan, H. & Macklis, J.D. Molecular logic of neocortical projection neuron specification, development and diversity. Nature reviews. Neuroscience 14, 755-769 (2013).

21. Lodato, S. & Arlotta, P. Generating neuronal diversity in the mammalian cerebral cortex. Annual review of cell and developmental biology 31, 699-720 (2015).

22. Kriegstein, A. & Alvarez-Buylla, A. The glial nature of embryonic and adult neural stem cells. Annual review of neuroscience 32, 149-184 (2009).

23. Hohmann, C. Cortical Development : From Specification to Differentiation. (Springer Berlin Heidelberg : Imprint : Springer, Berlin, Heidelberg; 2002).

24. Langlois, S.D., Morin, S., Yam, P.T. & Charron, F. Dissection and culture of commissural neurons from embryonic spinal cord. Journal of visualized experiments : JoVE (2010).

25. Naumann, R.K. et al. The reptilian brain. Current biology : CB 25, R317-321 (2015).

26. Glickstein, M. What does the cerebellum really do? Current biology : CB 17, R824-827 (2007).

27. Yuste, R., MacLean, J.N., Smith, J. & Lansner, A. The cortex as a central pattern generator. Nature reviews. Neuroscience 6, 477-483 (2005).

146

28. Toma, K. & Hanashima, C. Switching modes in corticogenesis: mechanisms of neuronal subtype transitions and integration in the cerebral cortex. Frontiers in neuroscience 9, 274 (2015).

29. Molyneaux, B.J., Arlotta, P., Menezes, J.R. & Macklis, J.D. Neuronal subtype specification in the cerebral cortex. Nature reviews. Neuroscience 8, 427-437 (2007).

30. Rakic, P., Ayoub, A.E., Breunig, J.J. & Dominguez, M.H. Decision by division: making cortical maps. Trends in neurosciences 32, 291-301 (2009).

31. O'Leary, D.D. & Nakagawa, Y. Patterning centers, regulatory genes and extrinsic mechanisms controlling arealization of the neocortex. Current opinion in neurobiology 12, 14-25 (2002).

32. Rakic, P. Specification of cerebral cortical areas. Science 241, 170-176 (1988).

33. Passingham, R.E., Stephan, K.E. & Kotter, R. The anatomical basis of functional localization in the cortex. Nature reviews. Neuroscience 3, 606-616 (2002).

36. DeFelipe, J. & DeFelipe, J. Cajal's neuronal forest : science and art.

37. Migliore, M. & Shepherd, G.M. Opinion: an integrated approach to classifying neuronal phenotypes. Nature reviews. Neuroscience 6, 810-818 (2005).

38. Johnson, M.B. & Walsh, C.A. Cerebral cortical neuron diversity and development at single-cell resolution. Current opinion in neurobiology 42, 9-16 (2017).

39. Aboitiz, F. & Montiel, J. One hundred million years of interhemispheric communication: the history of the corpus callosum. Brazilian journal of medical and biological research = Revista brasileira de pesquisas medicas e biologicas / Sociedade Brasileira de Biofisica ... [et al.] 36, 409-420 (2003).

40. Bandler, R.C., Mayer, C. & Fishell, G. Cortical interneuron specification: the juncture of genes, time and geometry. Current opinion in neurobiology 42, 17-24 (2017).

41. Marin, O. Interneuron dysfunction in psychiatric disorders. Nature reviews. Neuroscience 13, 107-120 (2012).

42. Markram, H. et al. Interneurons of the neocortical inhibitory system. Nature reviews. Neuroscience 5, 793-807 (2004).

147

43. Hemmati-Brivanlou, A. & Melton, D. Vertebrate embryonic cells will become nerve cells unless told otherwise. Cell 88, 13-17 (1997).

44. Munoz-Sanjuan, I. & Brivanlou, A.H. Neural induction, the default model and embryonic stem cells. Nature reviews. Neuroscience 3, 271-280 (2002).

45. Spemann, H. & Mangold, H. Induction of embryonic primordia by implantation of organizers from a different species. 1923. The International journal of developmental biology 45, 13-38 (2001).

46. Mangold, O. & Spemann, H. Wilhelm Roux' Archiv fur Entwicklungsmechanik der Organismen 111, 341-422 (1927).

47. Gilbert, S.F. Developmental biology, Edn. 8th. (Sinauer Associates, Inc. Publishers, Sunderland, Mass.; 2006).

48. Briscoe, J. & Small, S. Morphogen rules: design principles of gradient-mediated embryo patterning. Development 142, 3996-4009 (2015).

49. Tiberi, L., Vanderhaeghen, P. & van den Ameele, J. Cortical neurogenesis and morphogens: diversity of cues, sources and functions. Current opinion in cell biology 24, 269-276 (2012).

50. Rakic, P. Evolution of the neocortex: a perspective from developmental biology. Nature reviews. Neuroscience 10, 724-735 (2009).

51. Kriegstein, A., Noctor, S. & Martinez-Cerdeno, V. Patterns of neural stem and progenitor cell division may underlie evolutionary cortical expansion. Nature reviews. Neuroscience 7, 883-890 (2006).

52. Lodato, S., Shetty, A.S. & Arlotta, P. Cerebral cortex assembly: generating and reprogramming projection neuron diversity. Trends in neurosciences 38, 117-125 (2015).

53. Marin, O. & Rubenstein, J.L. A long, remarkable journey: tangential migration in the telencephalon. Nature reviews. Neuroscience 2, 780-790 (2001).

54. Molyneaux, B.J., Arlotta, P., Hirata, T., Hibi, M. & Macklis, J.D. Fezl is required for the birth and specification of corticospinal motor neurons. Neuron 47, 817-831 (2005).

55. Arlotta, P. et al. Neuronal subtype-specific genes that control corticospinal motor neuron development in vivo. Neuron 45, 207-221 (2005).

56. Britanova, O. et al. Satb2 is a postmitotic determinant for upper-layer neuron specification in the neocortex. Neuron 57, 378-392 (2008).

148

57. Leone, D.P. et al. Satb2 Regulates the Differentiation of Both Callosal and Subcerebral Projection Neurons in the Developing Cerebral Cortex. Cerebral cortex 25, 3406-3419 (2015).

58. Chen, B. et al. The Fezf2-Ctip2 genetic pathway regulates the fate choice of subcortical projection neurons in the developing cerebral cortex. Proceedings of the National Academy of Sciences of the United States of America 105, 11382-11387 (2008).

59. Alcamo, E.A. et al. Satb2 regulates callosal projection neuron identity in the developing cerebral cortex. Neuron 57, 364-377 (2008).

60. Srivatsa, S. et al. Unc5C and DCC act downstream of Ctip2 and Satb2 and contribute to corpus callosum formation. Nature communications 5, 3708 (2014).

61. Lai, T. et al. SOX5 controls the sequential generation of distinct corticofugal neuron subtypes. Neuron 57, 232-247 (2008).

62. Banker, G. The Development of Neuronal Polarity: A Retrospective View. The Journal of neuroscience : the official journal of the Society for Neuroscience 38, 1867-1873 (2018).

63. Kapitein, L.C. & Hoogenraad, C.C. Building the Neuronal Microtubule Cytoskeleton. Neuron 87, 492-506 (2015).

64. van Beuningen, S.F. & Hoogenraad, C.C. Neuronal polarity: remodeling microtubule organization. Current opinion in neurobiology 39, 1-7 (2016).

65. Tortosa, E. & Hoogenraad, C.C. Polarized trafficking: the palmitoylation cycle distributes cytoplasmic proteins to distinct neuronal compartments. Current opinion in cell biology 50, 64-71 (2018).

66. Schelski, M. & Bradke, F. Neuronal polarization: From spatiotemporal signaling to cytoskeletal dynamics. Molecular and cellular neurosciences 84, 11-28 (2017).

67. Dotti, C.G., Sullivan, C.A. & Banker, G.A. The establishment of polarity by hippocampal neurons in culture. The Journal of neuroscience : the official journal of the Society for Neuroscience 8, 1454-1468 (1988).

68. Cheng, P.L. & Poo, M.M. Early events in axon/dendrite polarization. Annual review of neuroscience 35, 181-201 (2012).

69. Winans, A.M., Collins, S.R. & Meyer, T. Waves of actin and microtubule polymerization drive microtubule-based transport and neurite growth before single axon formation. eLife 5, e12387 (2016).

149

70. Witte, H. & Bradke, F. The role of the cytoskeleton during neuronal polarization. Current opinion in neurobiology 18, 479-487 (2008).

71. Leterrier, C. The Axon Initial Segment, 50Years Later: A Nexus for Neuronal Organization and Function. Current topics in membranes 77, 185-233 (2016).

72. Huang, Y.M. & Rasband, M.N. Organization of the axon initial segment: Actin like a fence. The Journal of cell biology 215, 9-11 (2016).

73. Yogev, S. & Shen, K. Establishing Neuronal Polarity with Environmental and Intrinsic Mechanisms. Neuron 96, 638-650 (2017).

74. Funahashi, Y., Namba, T., Nakamuta, S. & Kaibuchi, K. Neuronal polarization in vivo: Growing in a complex environment. Current opinion in neurobiology 27, 215-223 (2014).

75. Leterrier, C., Dubey, P. & Roy, S. The nano-architecture of the axonal cytoskeleton. Nature reviews. Neuroscience 18, 713-726 (2017).

76. van Beuningen, S.F. et al. TRIM46 Controls Neuronal Polarity and Axon Specification by Driving the Formation of Parallel Microtubule Arrays. Neuron 88, 1208-1226 (2015).

77. Xu, K., Zhong, G. & Zhuang, X. Actin, spectrin, and associated proteins form a periodic cytoskeletal structure in axons. Science 339, 452-456 (2013).

78. Menager, C., Arimura, N., Fukata, Y. & Kaibuchi, K. PIP3 is involved in neuronal polarization and axon formation. Journal of neurochemistry 89, 109-118 (2004).

79. Jiang, H., Guo, W., Liang, X. & Rao, Y. Both the establishment and the maintenance of neuronal polarity require active mechanisms: critical roles of GSK-3beta and its upstream regulators. Cell 120, 123-135 (2005).

80. Shi, S.H., Jan, L.Y. & Jan, Y.N. Hippocampal neuronal polarity specified by spatially localized mPar3/mPar6 and PI 3-kinase activity. Cell 112, 63-75 (2003).

81. Mori, T., Wada, T., Suzuki, T., Kubota, Y. & Inagaki, N. Singar1, a novel RUN domain- containing protein, suppresses formation of surplus axons for neuronal polarity. The Journal of biological chemistry 282, 19884-19893 (2007).

82. Toriyama, M. et al. Shootin1: A protein involved in the organization of an asymmetric signal for neuronal polarization. The Journal of cell biology 175, 147-157 (2006).

83. Yoshimura, T. et al. Ras regulates neuronal polarity via the PI3-kinase/Akt/GSK- 3beta/CRMP-2 pathway. Biochemical and biophysical research communications 340, 62- 68 (2006).

150

84. Guo, W., Jiang, H., Gray, V., Dedhar, S. & Rao, Y. Role of the integrin-linked kinase (ILK) in determining neuronal polarity. Developmental biology 306, 457-468 (2007).

85. Gartner, A., Huang, X. & Hall, A. Neuronal polarity is regulated by glycogen synthase kinase-3 (GSK-3beta) independently of Akt/PKB serine phosphorylation. Journal of cell science 119, 3927-3934 (2006).

86. Yoshimura, T. et al. GSK-3beta regulates phosphorylation of CRMP-2 and neuronal polarity. Cell 120, 137-149 (2005).

87. Sperber, B.R., Leight, S., Goedert, M. & Lee, V.M. Glycogen synthase kinase-3 beta phosphorylates tau protein at multiple sites in intact cells. Neuroscience letters 197, 149- 153 (1995).

88. Gonzalez-Billault, C. et al. Microtubule-associated protein 1B function during normal development, regeneration, and pathological conditions in the nervous system. Journal of neurobiology 58, 48-59 (2004).

89. Shi, S.H., Cheng, T., Jan, L.Y. & Jan, Y.N. APC and GSK-3beta are involved in mPar3 targeting to the nascent axon and establishment of neuronal polarity. Current biology : CB 14, 2025-2032 (2004).

90. Inagaki, N. et al. CRMP-2 induces axons in cultured hippocampal neurons. Nature neuroscience 4, 781-782 (2001).

91. Kawano, Y. et al. CRMP-2 is involved in kinesin-1-dependent transport of the Sra- 1/WAVE1 complex and axon formation. Molecular and cellular biology 25, 9920-9935 (2005).

92. Goldstein, B. & Macara, I.G. The PAR proteins: fundamental players in animal cell polarization. Developmental cell 13, 609-622 (2007).

93. Lizcano, J.M. et al. LKB1 is a master kinase that activates 13 kinases of the AMPK subfamily, including MARK/PAR-1. The EMBO journal 23, 833-843 (2004).

94. Sapkota, G.P. et al. Phosphorylation of the protein kinase mutated in Peutz-Jeghers cancer syndrome, LKB1/STK11, at Ser431 by p90(RSK) and cAMP-dependent protein kinase, but not its farnesylation at Cys(433), is essential for LKB1 to suppress cell vrowth. The Journal of biological chemistry 276, 19469-19482 (2001).

95. Shelly, M., Cancedda, L., Heilshorn, S., Sumbre, G. & Poo, M.M. LKB1/STRAD promotes axon initiation during neuronal polarization. Cell 129, 565-577 (2007).

96. Barnes, A.P. et al. LKB1 and SAD kinases define a pathway required for the polarization of cortical neurons. Cell 129, 549-563 (2007).

151

97. Guo, S. & Kemphues, K.J. par-1, a gene required for establishing polarity in C. elegans embryos, encodes a putative Ser/Thr kinase that is asymmetrically distributed. Cell 81, 611-620 (1995).

98. Yamada, M. et al. Insulin receptor substrate (IRS)-1 and IRS-2 are tyrosine- phosphorylated and associated with phosphatidylinositol 3-kinase in response to brain- derived neurotrophic factor in cultured cerebral cortical neurons. The Journal of biological chemistry 272, 30334-30339 (1997).

99. Tsai, J.W., Chen, Y., Kriegstein, A.R. & Vallee, R.B. LIS1 RNA interference blocks neural stem cell division, morphogenesis, and motility at multiple stages. The Journal of cell biology 170, 935-945 (2005).

100. Nagano, T., Morikubo, S. & Sato, M. Filamin A and FILIP (Filamin A-Interacting Protein) regulate cell polarity and motility in neocortical subventricular and intermediate zones during radial migration. The Journal of neuroscience : the official journal of the Society for Neuroscience 24, 9648-9657 (2004).

101. Jossin, Y. & Cooper, J.A. Reelin, Rap1 and N-cadherin orient the migration of multipolar neurons in the developing neocortex. Nature neuroscience 14, 697-703 (2011).

102. Pacary, E. et al. Proneural transcription factors regulate different steps of cortical neuron migration through Rnd-mediated inhibition of RhoA signaling. Neuron 69, 1069-1084 (2011).

103. Kawauchi, T., Chihama, K., Nabeshima, Y. & Hoshino, M. The in vivo roles of STEF/Tiam1, Rac1 and JNK in cortical neuronal migration. The EMBO journal 22, 4190-4201 (2003).

104. Hall, A. & Lalli, G. Rho and Ras GTPases in axon growth, guidance, and branching. Cold Spring Harbor perspectives in biology 2, a001818 (2010).

105. Jaffe, A.B. & Hall, A. Rho GTPases: biochemistry and biology. Annual review of cell and developmental biology 21, 247-269 (2005).

106. Schwamborn, J.C. & Puschel, A.W. The sequential activity of the GTPases Rap1B and Cdc42 determines neuronal polarity. Nature neuroscience 7, 923-929 (2004).

107. Sofroniew, M.V., Howe, C.L. & Mobley, W.C. Nerve growth factor signaling, neuroprotection, and neural repair. Annual review of neuroscience 24, 1217-1281 (2001).

108. Oinuma, I., Katoh, H. & Negishi, M. R-Ras controls axon specification upstream of glycogen synthase kinase-3beta through integrin-linked kinase. The Journal of biological chemistry 282, 303-318 (2007).

152

109. Fivaz, M., Bandara, S., Inoue, T. & Meyer, T. Robust neuronal symmetry breaking by Ras-triggered local positive feedback. Current biology : CB 18, 44-50 (2008).

110. Nobes, C.D. & Hall, A. Rho, rac, and cdc42 GTPases regulate the assembly of multimolecular focal complexes associated with actin stress fibers, lamellipodia, and filopodia. Cell 81, 53-62 (1995).

111. Garvalov, B.K. et al. Cdc42 regulates cofilin during the establishment of neuronal polarity. The Journal of neuroscience : the official journal of the Society for Neuroscience 27, 13117-13129 (2007).

112. Nishimura, T. et al. PAR-6-PAR-3 mediates Cdc42-induced Rac activation through the Rac GEFs STEF/Tiam1. Nature cell biology 7, 270-277 (2005).

113. Watabe-Uchida, M., John, K.A., Janas, J.A., Newey, S.E. & Van Aelst, L. The Rac activator DOCK7 regulates neuronal polarity through local phosphorylation of stathmin/Op18. Neuron 51, 727-739 (2006).

114. Zhang, H. & Macara, I.G. The PAR-6 polarity protein regulates dendritic spine morphogenesis through p190 RhoGAP and the Rho GTPase. Developmental cell 14, 216- 226 (2008).

115. Jiang, W. et al. p190A RhoGAP is a glycogen synthase kinase-3-beta substrate required for polarized cell migration. The Journal of biological chemistry 283, 20978-20988 (2008).

116. Matheson, S.F. et al. Distinct but overlapping functions for the closely related p190 RhoGAPs in neural development. Developmental neuroscience 28, 538-550 (2006).

117. Kon, E., Cossard, A. & Jossin, Y. Neuronal Polarity in the Embryonic Mammalian Cerebral Cortex. Frontiers in cellular neuroscience 11, 163 (2017).

118. Namba, T. et al. Pioneering axons regulate neuronal polarization in the developing cerebral cortex. Neuron 81, 814-829 (2014).

119. Zhang, X. et al. Dishevelled promotes axon differentiation by regulating atypical protein kinase C. Nature cell biology 9, 743-754 (2007).

120. Sosa, L. et al. IGF-1 receptor is essential for the establishment of hippocampal neuronal polarity. Nature neuroscience 9, 993-995 (2006).

121. Yi, J.J., Barnes, A.P., Hand, R., Polleux, F. & Ehlers, M.D. TGF-beta signaling specifies axons during brain development. Cell 142, 144-157 (2010).

122. Chen, G. et al. Semaphorin-3A guides radial migration of cortical neurons during development. Nature neuroscience 11, 36-44 (2008).

153

123. Zhang, J.H. et al. DCC-Mediated Dab1 Phosphorylation Participates in the Multipolar-to- Bipolar Transition of Migrating Neurons. Cell reports 22, 3598-3611 (2018).

124. Adler, C.E., Fetter, R.D. & Bargmann, C.I. UNC-6/Netrin induces neuronal asymmetry and defines the site of axon formation. Nature neuroscience 9, 511-518 (2006).

125. Esch, T., Lemmon, V. & Banker, G. Local presentation of substrate molecules directs axon specification by cultured hippocampal neurons. The Journal of neuroscience : the official journal of the Society for Neuroscience 19, 6417-6426 (1999).

126. Flynn, K.C. The cytoskeleton and neurite initiation. Bioarchitecture 3, 86-109 (2013).

127. Schaefer, A.W., Kabir, N. & Forscher, P. Filopodia and actin arcs guide the assembly and transport of two populations of microtubules with unique dynamic parameters in neuronal growth cones. The Journal of cell biology 158, 139-152 (2002).

128. Dent, E.W. & Gertler, F.B. Cytoskeletal dynamics and transport in growth cone motility and axon guidance. Neuron 40, 209-227 (2003).

129. Pak, C.W., Flynn, K.C. & Bamburg, J.R. Actin-binding proteins take the reins in growth cones. Nature reviews. Neuroscience 9, 136-147 (2008).

130. Coles, C.H. & Bradke, F. Coordinating neuronal actin-microtubule dynamics. Current biology : CB 25, R677-691 (2015).

131. Liu, G. & Dwyer, T. Microtubule dynamics in axon guidance. Neuroscience bulletin 30, 569-583 (2014).

132. Goldberg, D.J. & Burmeister, D.W. Stages in axon formation: observations of growth of Aplysia axons in culture using video-enhanced contrast-differential interference contrast microscopy. The Journal of cell biology 103, 1921-1931 (1986).

133. Alberts, B. Molecular biology of the cell, Edn. 5th. (Garland Science, New York; 2008).

134. Dent, E.W., Gupton, S.L. & Gertler, F.B. The growth cone cytoskeleton in axon outgrowth and guidance. Cold Spring Harbor perspectives in biology 3 (2011).

135. Krause, M., Dent, E.W., Bear, J.E., Loureiro, J.J. & Gertler, F.B. Ena/VASP proteins: regulators of the actin cytoskeleton and cell migration. Annual review of cell and developmental biology 19, 541-564 (2003).

136. Sarmiere, P.D. & Bamburg, J.R. Regulation of the neuronal actin cytoskeleton by ADF/cofilin. Journal of neurobiology 58, 103-117 (2004).

154

137. Ridley, A.J., Paterson, H.F., Johnston, C.L., Diekmann, D. & Hall, A. The small GTP- binding protein rac regulates growth factor-induced membrane ruffling. Cell 70, 401-410 (1992).

138. Ridley, A.J. et al. Cell migration: integrating signals from front to back. Science 302, 1704-1709 (2003).

139. Bard, L. et al. A molecular clutch between the actin flow and N-cadherin adhesions drives growth cone migration. The Journal of neuroscience : the official journal of the Society for Neuroscience 28, 5879-5890 (2008).

140. Mitchison, T. & Kirschner, M. Cytoskeletal dynamics and nerve growth. Neuron 1, 761- 772 (1988).

141. Medeiros, N.A., Burnette, D.T. & Forscher, P. Myosin II functions in actin-bundle turnover in neuronal growth cones. Nature cell biology 8, 215-226 (2006).

142. Marsh, L. & Letourneau, P.C. Growth of neurites without filopodial or lamellipodial activity in the presence of cytochalasin B. The Journal of cell biology 99, 2041-2047 (1984).

143. Bradke, F. & Dotti, C.G. The role of local actin instability in axon formation. Science 283, 1931-1934 (1999).

144. Zheng, J.Q., Wan, J.J. & Poo, M.M. Essential role of filopodia in chemotropic turning of nerve growth cone induced by a glutamate gradient. The Journal of neuroscience : the official journal of the Society for Neuroscience 16, 1140-1149 (1996).

145. Bentley, D. & Toroian-Raymond, A. Disoriented pathfinding by pioneer neurone growth cones deprived of filopodia by cytochalasin treatment. Nature 323, 712-715 (1986).

146. Dwivedy, A., Gertler, F.B., Miller, J., Holt, C.E. & Lebrand, C. Ena/VASP function in retinal axons is required for terminal arborization but not pathway navigation. Development 134, 2137-2146 (2007).

147. Conde, C. & Caceres, A. Microtubule assembly, organization and dynamics in axons and dendrites. Nature reviews. Neuroscience 10, 319-332 (2009).

148. Kahn, O.I. & Baas, P.W. Microtubules and Growth Cones: Motors Drive the Turn. Trends in neurosciences 39, 433-440 (2016).

149. Buck, K.B. & Zheng, J.Q. Growth cone turning induced by direct local modification of microtubule dynamics. The Journal of neuroscience : the official journal of the Society for Neuroscience 22, 9358-9367 (2002).

155

150. Sabry, J.H. et al. Microtubule behavior during guidance of pioneer neuron growth cones in situ. The Journal of cell biology 115, 381-395 (1991).

151. Tanaka, E. & Kirschner, M.W. The role of microtubules in growth cone turning at substrate boundaries. The Journal of cell biology 128, 127-137 (1995).

152. Tanaka, E., Ho, T. & Kirschner, M.W. The role of microtubule dynamics in growth cone motility and axonal growth. The Journal of cell biology 128, 139-155 (1995).

153. Mitchison, T.J. & Cramer, L.P. Actin-based cell motility and cell locomotion. Cell 84, 371-379 (1996).

154. Goldberg, D.J. & Burmeister, D.W. Looking into growth cones. Trends in neurosciences 12, 503-506 (1989).

155. Dent, E.W., Barnes, A.M., Tang, F. & Kalil, K. Netrin-1 and semaphorin 3A promote or inhibit cortical axon branching, respectively, by reorganization of the cytoskeleton. The Journal of neuroscience : the official journal of the Society for Neuroscience 24, 3002- 3012 (2004).

156. Lin, C.H. & Forscher, P. Cytoskeletal remodeling during growth cone-target interactions. The Journal of cell biology 121, 1369-1383 (1993).

157. Lin, C.H. & Forscher, P. Growth cone advance is inversely proportional to retrograde F- actin flow. Neuron 14, 763-771 (1995).

158. Nugent, A.A., Kolpak, A.L. & Engle, E.C. Human disorders of axon guidance. Current opinion in neurobiology 22, 837-843 (2012).

159. Poirier, K. et al. Mutations in the neuronal ss-tubulin subunit TUBB3 result in malformation of cortical development and neuronal migration defects. Human molecular genetics 19, 4462-4473 (2010).

160. Tischfield, M.A. et al. Human TUBB3 mutations perturb microtubule dynamics, kinesin interactions, and axon guidance. Cell 140, 74-87 (2010).

161. Purro, S.A. et al. Wnt regulates axon behavior through changes in microtubule growth directionality: a new role for adenomatous polyposis coli. The Journal of neuroscience : the official journal of the Society for Neuroscience 28, 8644-8654 (2008).

162. Suter, D.M., Schaefer, A.W. & Forscher, P. Microtubule dynamics are necessary for SRC family kinase-dependent growth cone steering. Current biology : CB 14, 1194-1199 (2004).

156

163. Sengottuvel, V., Leibinger, M., Pfreimer, M., Andreadaki, A. & Fischer, D. Taxol facilitates axon regeneration in the mature CNS. The Journal of neuroscience : the official journal of the Society for Neuroscience 31, 2688-2699 (2011).

164. Qu, C. et al. Direct binding of TUBB3 with DCC couples netrin-1 signaling to intracellular microtubule dynamics in axon outgrowth and guidance. Journal of cell science 126, 3070-3081 (2013).

165. Liu, G. et al. Netrin requires focal adhesion kinase and Src family kinases for axon outgrowth and attraction. Nature neuroscience 7, 1222-1232 (2004).

166. Li, W. et al. Activation of FAK and Src are receptor-proximal events required for netrin signaling. Nature neuroscience 7, 1213-1221 (2004).

167. Zhou, F.Q., Waterman-Storer, C.M. & Cohan, C.S. Focal loss of actin bundles causes microtubule redistribution and growth cone turning. The Journal of cell biology 157, 839- 849 (2002).

168. Lee, A.C. & Suter, D.M. Quantitative analysis of microtubule dynamics during adhesion- mediated growth cone guidance. Developmental neurobiology 68, 1363-1377 (2008).

169. Applewhite, D.A. et al. The spectraplakin Short stop is an actin-microtubule cross-linker that contributes to organization of the microtubule network. Molecular biology of the cell 21, 1714-1724 (2010).

170. Ka, M., Jung, E.M., Mueller, U. & Kim, W.Y. MACF1 regulates the migration of pyramidal neurons via microtubule dynamics and GSK-3 signaling. Developmental biology 395, 4-18 (2014).

171. Grabham, P.W., Seale, G.E., Bennecib, M., Goldberg, D.J. & Vallee, R.B. Cytoplasmic dynein and LIS1 are required for microtubule advance during growth cone remodeling and fast axonal outgrowth. The Journal of neuroscience : the official journal of the Society for Neuroscience 27, 5823-5834 (2007).

172. Hasaka, T.P., Myers, K.A. & Baas, P.W. Role of actin filaments in the axonal transport of microtubules. The Journal of neuroscience : the official journal of the Society for Neuroscience 24, 11291-11301 (2004).

173. Lin, W. & Szaro, B.G. Neurofilaments help maintain normal morphologies and support elongation of neurites in Xenopus laevis cultured embryonic spinal cord neurons. The Journal of neuroscience : the official journal of the Society for Neuroscience 15, 8331- 8344 (1995).

174. Eyer, J. & Peterson, A. Neurofilament-deficient axons and perikaryal aggregates in viable transgenic mice expressing a neurofilament-beta-galactosidase fusion protein. Neuron 12, 389-405 (1994).

157

175. Laser-Azogui, A., Kornreich, M., Malka-Gibor, E. & Beck, R. Neurofilament assembly and function during neuronal development. Current opinion in cell biology 32, 92-101 (2015).

176. Markus, A., Zhong, J. & Snider, W.D. Raf and akt mediate distinct aspects of sensory axon growth. Neuron 35, 65-76 (2002).

177. Uesugi, K., Oinuma, I., Katoh, H. & Negishi, M. Different requirement for Rnd GTPases of R-Ras GAP activity of Plexin-C1 and Plexin-D1. The Journal of biological chemistry 284, 6743-6751 (2009).

178. Oinuma, I., Ishikawa, Y., Katoh, H. & Negishi, M. The Semaphorin 4D receptor Plexin- B1 is a GTPase activating protein for R-Ras. Science 305, 862-865 (2004).

179. Richter, M., Murai, K.K., Bourgin, C., Pak, D.T. & Pasquale, E.B. The EphA4 receptor regulates neuronal morphology through SPAR-mediated inactivation of Rap GTPases. The Journal of neuroscience : the official journal of the Society for Neuroscience 27, 14205-14215 (2007).

180. Antoine-Bertrand, J. et al. p120RasGAP Protein Mediates Netrin-1 Protein-induced Cortical Axon Outgrowth and Guidance. The Journal of biological chemistry 291, 4589- 4602 (2016).

181. Barberis, D. et al. p190 Rho-GTPase activating protein associates with plexins and it is required for semaphorin signalling. Journal of cell science 118, 4689-4700 (2005).

182. Iwasato, T. et al. Rac-GAP alpha-chimerin regulates motor-circuit formation as a key mediator of EphrinB3/EphA4 forward signaling. Cell 130, 742-753 (2007).

183. Briancon-Marjollet, A. et al. Trio mediates netrin-1-induced Rac1 activation in axon outgrowth and guidance. Molecular and cellular biology 28, 2314-2323 (2008).

184. Forsthoefel, D.J., Liebl, E.C., Kolodziej, P.A. & Seeger, M.A. The Abelson tyrosine kinase, the Trio GEF and Enabled interact with the Netrin receptor Frazzled in Drosophila. Development 132, 1983-1994 (2005).

185. Li, X. et al. Netrin signal transduction and the guanine nucleotide exchange factor DOCK180 in attractive signaling. Nature neuroscience 11, 28-35 (2008).

186. Bashaw, G.J. & Klein, R. Signaling from axon guidance receptors. Cold Spring Harbor perspectives in biology 2, a001941 (2010).

187. O'Donnell, M., Chance, R.K. & Bashaw, G.J. Axon growth and guidance: receptor regulation and signal transduction. Annual review of neuroscience 32, 383-412 (2009).

158

188. Stoeckli, E.T. Understanding axon guidance: are we nearly there yet? Development 145 (2018).

189. Neuhaus-Follini, A. & Bashaw, G.J. Crossing the embryonic midline: molecular mechanisms regulating axon responsiveness at an intermediate target. Wiley interdisciplinary reviews. Developmental biology 4, 377-389 (2015).

190. Tessier-Lavigne, M. & Goodman, C.S. The molecular biology of axon guidance. Science 274, 1123-1133 (1996).

191. de Ramon Francas, G., Zuniga, N.R. & Stoeckli, E.T. The spinal cord shows the way - How axons navigate intermediate targets. Developmental biology 432, 43-52 (2017).

192. Kolodkin, A.L. & Tessier-Lavigne, M. Mechanisms and molecules of neuronal wiring: a primer. Cold Spring Harbor perspectives in biology 3 (2011).

193. Charron, F. & Tessier-Lavigne, M. Novel brain wiring functions for classical morphogens: a role as graded positional cues in axon guidance. Development 132, 2251- 2262 (2005).

194. Kennedy, T.E., Serafini, T., de la Torre, J.R. & Tessier-Lavigne, M. Netrins are diffusible chemotropic factors for commissural axons in the embryonic spinal cord. Cell 78, 425- 435 (1994).

195. Keino-Masu, K. et al. Deleted in Colorectal Cancer (DCC) encodes a netrin receptor. Cell 87, 175-185 (1996).

196. Lai Wing Sun, K., Correia, J.P. & Kennedy, T.E. Netrins: versatile extracellular cues with diverse functions. Development 138, 2153-2169 (2011).

197. Blockus, H. & Chedotal, A. Slit-Robo signaling. Development 143, 3037-3044 (2016).

198. Stein, E. & Tessier-Lavigne, M. Hierarchical organization of guidance receptors: silencing of netrin attraction by slit through a Robo/DCC receptor complex. Science 291, 1928-1938 (2001).

199. Nawabi, H. et al. A midline switch of receptor processing regulates commissural axon guidance in vertebrates. Genes & development 24, 396-410 (2010).

200. Kadison, S.R., Makinen, T., Klein, R., Henkemeyer, M. & Kaprielian, Z. EphB receptors and ephrin-B3 regulate axon guidance at the ventral midline of the embryonic mouse spinal cord. The Journal of neuroscience : the official journal of the Society for Neuroscience 26, 8909-8914 (2006).

159

201. Morales, D. & Kania, A. Cooperation and crosstalk in axon guidance cue integration: Additivity, synergy, and fine-tuning in combinatorial signaling. Developmental neurobiology 77, 891-904 (2017).

202. Jessell, T.M. Neuronal specification in the spinal cord: inductive signals and transcriptional codes. Nature reviews. Genetics 1, 20-29 (2000).

203. Charron, F., Stein, E., Jeong, J., McMahon, A.P. & Tessier-Lavigne, M. The morphogen sonic hedgehog is an axonal chemoattractant that collaborates with netrin-1 in midline axon guidance. Cell 113, 11-23 (2003).

204. Ruiz de Almodovar, C. et al. VEGF mediates commissural axon chemoattraction through its receptor Flk1. Neuron 70, 966-978 (2011).

205. Sloan, T.F., Qasaimeh, M.A., Juncker, D., Yam, P.T. & Charron, F. Integration of shallow gradients of Shh and Netrin-1 guides commissural axons. PLoS biology 13, e1002119 (2015).

206. Meriane, M. et al. Phosphorylation of DCC by Fyn mediates Netrin-1 signaling in growth cone guidance. The Journal of cell biology 167, 687-698 (2004).

207. Yam, P.T. et al. 14-3-3 proteins regulate a cell-intrinsic switch from sonic hedgehog- mediated commissural axon attraction to repulsion after midline crossing. Neuron 76, 735-749 (2012).

208. Butler, S.J. & Dodd, J. A role for BMP heterodimers in roof plate-mediated repulsion of commissural axons. Neuron 38, 389-401 (2003).

209. Islam, S.M. et al. Draxin, a repulsive guidance protein for spinal cord and forebrain commissures. Science 323, 388-393 (2009).

210. Lyuksyutova, A.I. et al. Anterior-posterior guidance of commissural axons by Wnt- frizzled signaling. Science 302, 1984-1988 (2003).

211. Dickson, B.J. Wnts send axons up and down the spinal cord. Nature neuroscience 8, 1130-1132 (2005).

212. Rothberg, J.M., Jacobs, J.R., Goodman, C.S. & Artavanis-Tsakonas, S. slit: an extracellular protein necessary for development of midline glia and commissural axon pathways contains both EGF and LRR domains. Genes & development 4, 2169-2187 (1990).

213. Rothberg, J.M., Hartley, D.A., Walther, Z. & Artavanis-Tsakonas, S. slit: an EGF- homologous locus of D. melanogaster involved in the development of the embryonic central nervous system. Cell 55, 1047-1059 (1988).

160

214. Nusslein-Volhard, C., Wieschaus, E. & Kluding, H. Mutations affecting the pattern of the larval cuticle inDrosophila melanogaster : I. Zygotic loci on the second chromosome. Wilhelm Roux's archives of developmental biology 193, 267-282 (1984).

215. Kidd, T. et al. Roundabout controls axon crossing of the CNS midline and defines a novel subfamily of evolutionarily conserved guidance receptors. Cell 92, 205-215 (1998).

216. Seeger, M., Tear, G., Ferres-Marco, D. & Goodman, C.S. Mutations affecting growth cone guidance in Drosophila: genes necessary for guidance toward or away from the midline. Neuron 10, 409-426 (1993).

217. Tear, G., Seeger, M. & Goodman, C.S. To cross or not to cross: a genetic analysis of guidance at the midline. Perspectives on developmental neurobiology 1, 183-194 (1993).

218. Zelina, P. et al. Signaling switch of the axon guidance receptor Robo3 during vertebrate evolution. Neuron 84, 1258-1272 (2014).

219. Bashaw, G.J., Kidd, T., Murray, D., Pawson, T. & Goodman, C.S. Repulsive axon guidance: Abelson and Enabled play opposing roles downstream of the roundabout receptor. Cell 101, 703-715 (2000).

220. Wong, K. et al. Signal transduction in neuronal migration: roles of GTPase activating proteins and the small GTPase Cdc42 in the Slit-Robo pathway. Cell 107, 209-221 (2001).

221. Round, J.E. & Sun, H. The adaptor protein Nck2 mediates Slit1-induced changes in cortical neuron morphology. Molecular and cellular neurosciences 47, 265-273 (2011).

222. Yang, L. & Bashaw, G.J. Son of sevenless directly links the Robo receptor to rac activation to control axon repulsion at the midline. Neuron 52, 595-607 (2006).

223. Guerrier, S. et al. The F-BAR domain of srGAP2 induces membrane protrusions required for neuronal migration and morphogenesis. Cell 138, 990-1004 (2009).

224. Charrier, C. et al. Inhibition of SRGAP2 function by its human-specific paralogs induces neoteny during spine maturation. Cell 149, 923-935 (2012).

225. Georgiou, M. & Tear, G. Commissureless is required both in commissural neurones and midline cells for axon guidance across the midline. Development 129, 2947-2956 (2002).

226. Keleman, K. et al. Comm sorts robo to control axon guidance at the Drosophila midline. Cell 110, 415-427 (2002).

227. Yazdani, U. & Terman, J.R. The semaphorins. Genome biology 7, 211 (2006).

161

228. Jongbloets, B.C. & Pasterkamp, R.J. Semaphorin signalling during development. Development 141, 3292-3297 (2014).

229. Hota, P.K. & Buck, M. Plexin structures are coming: opportunities for multilevel investigations of semaphorin guidance receptors, their cell signaling mechanisms, and functions. Cellular and molecular life sciences : CMLS 69, 3765-3805 (2012).

230. Fujisawa, H. Discovery of semaphorin receptors, neuropilin and plexin, and their functions in neural development. Journal of neurobiology 59, 24-33 (2004).

231. Yang, T. & Terman, J.R. 14-3-3epsilon couples protein kinase A to semaphorin signaling and silences plexin RasGAP-mediated axonal repulsion. Neuron 74, 108-121 (2012).

232. Hung, R.J. et al. Mical links semaphorins to F-actin disassembly. Nature 463, 823-827 (2010).

233. Aizawa, H. et al. Phosphorylation of cofilin by LIM-kinase is necessary for semaphorin 3A-induced growth cone collapse. Nature neuroscience 4, 367-373 (2001).

234. Bellon, A. et al. VEGFR2 (KDR/Flk1) signaling mediates axon growth in response to semaphorin 3E in the developing brain. Neuron 66, 205-219 (2010).

235. Sperry, R.W. Chemoaffinity in the Orderly Growth of Nerve Fiber Patterns and Connections. Proceedings of the National Academy of Sciences of the United States of America 50, 703-710 (1963).

236. Attardi, D.G. & Sperry, R.W. Preferential selection of central pathways by regenerating optic fibers. Experimental neurology 7, 46-64 (1963).

237. Cheng, H.J., Nakamoto, M., Bergemann, A.D. & Flanagan, J.G. Complementary gradients in expression and binding of ELF-1 and Mek4 in development of the topographic retinotectal projection map. Cell 82, 371-381 (1995).

238. Drescher, U. et al. In vitro guidance of retinal ganglion cell axons by RAGS, a 25 kDa tectal protein related to ligands for Eph receptor tyrosine kinases. Cell 82, 359-370 (1995).

239. Yokoyama, N. et al. Forward signaling mediated by ephrin-B3 prevents contralateral corticospinal axons from recrossing the spinal cord midline. Neuron 29, 85-97 (2001).

240. Kullander, K. et al. Ephrin-B3 is the midline barrier that prevents corticospinal tract axons from recrossing, allowing for unilateral motor control. Genes & development 15, 877-888 (2001).

162

241. Hu, Z. et al. Corpus callosum deficiency in transgenic mice expressing a truncated ephrin-A receptor. The Journal of neuroscience : the official journal of the Society for Neuroscience 23, 10963-10970 (2003).

242. Mire, E. et al. Developmental Upregulation of Ephrin-B1 Silences Sema3C/Neuropilin-1 Signaling during Post-crossing Navigation of Corpus Callosum Axons. Current biology : CB 28, 1768-1782 e1764 (2018).

243. Cramer, K.S. & Miko, I.J. Eph-ephrin signaling in nervous system development. F1000Research 5 (2016).

244. Lisabeth, E.M., Falivelli, G. & Pasquale, E.B. Eph receptor signaling and ephrins. Cold Spring Harbor perspectives in biology 5 (2013).

245. Klein, R. Bidirectional modulation of synaptic functions by Eph/ephrin signaling. Nature neuroscience 12, 15-20 (2009).

246. Shamah, S.M. et al. EphA receptors regulate growth cone dynamics through the novel guanine nucleotide exchange factor ephexin. Cell 105, 233-244 (2001).

247. Haupaix, N. et al. p120RasGAP mediates ephrin/Eph-dependent attenuation of FGF/ERK signals during cell fate specification in ascidian embryos. Development 140, 4347-4352 (2013).

248. Dail, M., Richter, M., Godement, P. & Pasquale, E.B. Eph receptors inactivate R-Ras through different mechanisms to achieve cell repulsion. Journal of cell science 119, 1244-1254 (2006).

249. Cowan, C.A. & Henkemeyer, M. The SH2/SH3 adaptor Grb4 transduces B-ephrin reverse signals. Nature 413, 174-179 (2001).

250. Knoll, B. & Drescher, U. Src family kinases are involved in EphA receptor-mediated retinal axon guidance. The Journal of neuroscience : the official journal of the Society for Neuroscience 24, 6248-6257 (2004).

251. Serafini, T. et al. The netrins define a family of axon outgrowth-promoting proteins homologous to C. elegans UNC-6. Cell 78, 409-424 (1994).

252. Ishii, N., Wadsworth, W.G., Stern, B.D., Culotti, J.G. & Hedgecock, E.M. UNC-6, a laminin-related protein, guides cell and pioneer axon migrations in C. elegans. Neuron 9, 873-881 (1992).

253. Chan, S.S. et al. UNC-40, a C. elegans homolog of DCC (Deleted in Colorectal Cancer), is required in motile cells responding to UNC-6 netrin cues. Cell 87, 187-195 (1996).

163

254. Hong, K. et al. A ligand-gated association between cytoplasmic domains of UNC5 and DCC family receptors converts netrin-induced growth cone attraction to repulsion. Cell 97, 927-941 (1999).

255. Dominici, C. et al. Floor-plate-derived netrin-1 is dispensable for commissural axon guidance. Nature 545, 350-354 (2017).

256. Varadarajan, S.G. et al. Netrin1 Produced by Neural Progenitors, Not Floor Plate Cells, Is Required for Axon Guidance in the Spinal Cord. Neuron 94, 790-799 e793 (2017).

257. Fearon, E.R. et al. Identification of a chromosome 18q gene that is altered in colorectal cancers. Science 247, 49-56 (1990).

258. Deiner, M.S. et al. Netrin-1 and DCC mediate axon guidance locally at the optic disc: loss of function leads to optic nerve hypoplasia. Neuron 19, 575-589 (1997).

259. Seiradake, E., Jones, E.Y. & Klein, R. Structural Perspectives on Axon Guidance. Annual review of cell and developmental biology 32, 577-608 (2016).

260. Xu, K. et al. Neural migration. Structures of netrin-1 bound to two receptors provide insight into its axon guidance mechanism. Science 344, 1275-1279 (2014).

261. Finci, L.I. et al. The crystal structure of netrin-1 in complex with DCC reveals the bifunctionality of netrin-1 as a guidance cue. Neuron 83, 839-849 (2014).

262. Chen, Q. et al. N-terminal horseshoe conformation of DCC is functionally required for axon guidance and might be shared by other neural receptors. Journal of cell science 126, 186-195 (2013).

263. Finci, L., Zhang, Y., Meijers, R. & Wang, J.H. Signaling mechanism of the netrin-1 receptor DCC in axon guidance. Progress in biophysics and molecular biology 118, 153- 160 (2015).

264. Stein, E., Zou, Y., Poo, M. & Tessier-Lavigne, M. Binding of DCC by netrin-1 to mediate axon guidance independent of adenosine A2B receptor activation. Science 291, 1976-1982 (2001).

265. Ren, X.R. et al. Focal adhesion kinase in netrin-1 signaling. Nature neuroscience 7, 1204-1212 (2004).

266. Li, X. et al. The adaptor protein Nck-1 couples the netrin-1 receptor DCC (deleted in colorectal cancer) to the activation of the small GTPase Rac1 through an atypical mechanism. The Journal of biological chemistry 277, 37788-37797 (2002).

164

267. Lane, C., Qi, J. & Fawcett, J.P. NCK is critical for the development of deleted in colorectal cancer (DCC) sensitive spinal circuits. Journal of neurochemistry 134, 1008- 1014 (2015).

268. Zhu, X.J. et al. Myosin X regulates netrin receptors and functions in axonal path-finding. Nature cell biology 9, 184-192 (2007).

269. Shekarabi, M. et al. Deleted in colorectal cancer binding netrin-1 mediates cell substrate adhesion and recruits Cdc42, Rac1, Pak1, and N-WASP into an intracellular signaling complex that promotes growth cone expansion. The Journal of neuroscience : the official journal of the Society for Neuroscience 25, 3132-3141 (2005).

270. Shekarabi, M. & Kennedy, T.E. The netrin-1 receptor DCC promotes filopodia formation and cell spreading by activating Cdc42 and Rac1. Molecular and cellular neurosciences 19, 1-17 (2002).

271. Li, X., Saint-Cyr-Proulx, E., Aktories, K. & Lamarche-Vane, N. Rac1 and Cdc42 but not RhoA or Rho kinase activities are required for neurite outgrowth induced by the Netrin-1 receptor DCC (deleted in colorectal cancer) in N1E-115 neuroblastoma cells. The Journal of biological chemistry 277, 15207-15214 (2002).

272. DeGeer, J. et al. Hsc70 chaperone activity underlies Trio GEF function in axon growth and guidance induced by netrin-1. The Journal of cell biology 210, 817-832 (2015).

273. Antoine-Bertrand, J., Ghogha, A., Luangrath, V., Bedford, F.K. & Lamarche-Vane, N. The activation of ezrin-radixin-moesin proteins is regulated by netrin-1 through Src kinase and RhoA/Rho kinase activities and mediates netrin-1-induced axon outgrowth. Molecular biology of the cell 22, 3734-3746 (2011).

274. Deming, P.B. et al. Anchoring of protein kinase A by ERM (ezrin-radixin-moesin) proteins is required for proper netrin signaling through DCC (deleted in colorectal cancer). The Journal of biological chemistry 290, 5783-5796 (2015).

275. Menon, S. et al. The E3 Ubiquitin Ligase TRIM9 Is a Filopodia Off Switch Required for Netrin-Dependent Axon Guidance. Developmental cell 35, 698-712 (2015).

276. Plooster, M. et al. TRIM9-dependent ubiquitination of DCC constrains kinase signaling, exocytosis, and axon branching. Molecular biology of the cell 28, 2374-2385 (2017).

277. Leyva-Diaz, E. et al. FLRT3 is a Robo1-interacting protein that determines Netrin-1 attraction in developing axons. Current biology : CB 24, 494-508 (2014).

278. Ma, W. et al. Phosphorylation of DCC by ERK2 is facilitated by direct docking of the receptor P1 domain to the kinase. Structure 18, 1502-1511 (2010).

165

279. Forcet, C. et al. Netrin-1-mediated axon outgrowth requires deleted in colorectal cancer- dependent MAPK activation. Nature 417, 443-447 (2002).

280. Campbell, D.S. & Holt, C.E. Chemotropic responses of retinal growth cones mediated by rapid local protein synthesis and degradation. Neuron 32, 1013-1026 (2001).

281. Xie, Y. et al. DCC-dependent phospholipase C signaling in netrin-1-induced neurite elongation. The Journal of biological chemistry 281, 2605-2611 (2006).

282. Xie, Y. et al. Phosphatidylinositol transfer protein-alpha in netrin-1-induced PLC signalling and neurite outgrowth. Nature cell biology 7, 1124-1132 (2005).

283. Wang, G.X. & Poo, M.M. Requirement of TRPC channels in netrin-1-induced chemotropic turning of nerve growth cones. Nature 434, 898-904 (2005).

284. Shim, S. et al. XTRPC1-dependent chemotropic guidance of neuronal growth cones. Nature neuroscience 8, 730-735 (2005).

285. Clapham, D.E. Calcium signaling. Cell 131, 1047-1058 (2007).

286. Welshhans, K. & Bassell, G.J. Netrin-1-induced local beta-actin synthesis and growth cone guidance requires zipcode binding protein 1. The Journal of neuroscience : the official journal of the Society for Neuroscience 31, 9800-9813 (2011).

287. Leung, K.M. et al. Asymmetrical beta-actin mRNA translation in growth cones mediates attractive turning to netrin-1. Nature neuroscience 9, 1247-1256 (2006).

288. Tcherkezian, J., Brittis, P.A., Thomas, F., Roux, P.P. & Flanagan, J.G. Transmembrane receptor DCC associates with protein synthesis machinery and regulates translation. Cell 141, 632-644 (2010).

289. Campbell, D.S. & Holt, C.E. Apoptotic pathway and MAPKs differentially regulate chemotropic responses of retinal growth cones. Neuron 37, 939-952 (2003).

290. Finger, J.H. et al. The netrin 1 receptors Unc5h3 and Dcc are necessary at multiple choice points for the guidance of corticospinal tract axons. The Journal of neuroscience : the official journal of the Society for Neuroscience 22, 10346-10356 (2002).

291. Srour, M. et al. Mutations in DCC cause congenital mirror movements. Science 328, 592 (2010).

292. Meneret, A. et al. Mutations in the netrin-1 gene cause congenital mirror movements. The Journal of clinical investigation 127, 3923-3936 (2017).

166

293. Marsh, A.P.L. et al. DCC mutation update: Congenital mirror movements, isolated agenesis of the corpus callosum, and developmental split brain syndrome. Human mutation 39, 23-39 (2018).

294. Bierhals, T. et al. Novel DCC variants in congenital mirror movements and evaluation of disease-associated missense variants. European journal of medical genetics 61, 329-334 (2018).

295. Marsh, A.P. et al. Mutations in DCC cause isolated agenesis of the corpus callosum with incomplete penetrance. Nature genetics 49, 511-514 (2017).

296. da Silva, R.V. et al. DCC Is Required for the Development of Nociceptive Topognosis in Mice and Humans. Cell reports 22, 1105-1114 (2018).

297. Hoops, D. & Flores, C. Making Dopamine Connections in Adolescence. Trends in neurosciences 40, 709-719 (2017).

298. Kim, J.M. et al. SNPs in axon guidance pathway genes and susceptibility for Parkinson's disease in the Korean population. Journal of human genetics 56, 125-129 (2011).

299. Taniguchi, Y., Kim, S.H. & Sisodia, S.S. Presenilin-dependent "gamma-secretase" processing of deleted in colorectal cancer (DCC). The Journal of biological chemistry 278, 30425-30428 (2003).

300. Parent, A.T., Barnes, N.Y., Taniguchi, Y., Thinakaran, G. & Sisodia, S.S. Presenilin attenuates receptor-mediated signaling and synaptic function. The Journal of neuroscience : the official journal of the Society for Neuroscience 25, 1540-1549 (2005).

301. Bai, G. et al. Presenilin-dependent receptor processing is required for axon guidance. Cell 144, 106-118 (2011).

302. Galko, M.J. & Tessier-Lavigne, M. Function of an axonal chemoattractant modulated by metalloprotease activity. Science 289, 1365-1367 (2000).

303. Winkle, C.C. et al. A novel Netrin-1-sensitive mechanism promotes local SNARE- mediated exocytosis during axon branching. The Journal of cell biology 205, 217-232 (2014).

304. Bouchard, J.F., Horn, K.E., Stroh, T. & Kennedy, T.E. Depolarization recruits DCC to the plasma membrane of embryonic cortical neurons and enhances axon extension in response to netrin-1. Journal of neurochemistry 107, 398-417 (2008).

305. Wu, L.G., Hamid, E., Shin, W. & Chiang, H.C. Exocytosis and endocytosis: modes, functions, and coupling mechanisms. Annual review of physiology 76, 301-331 (2014).

167

306. Bartoe, J.L. et al. Protein interacting with C-kinase 1/protein kinase Calpha-mediated endocytosis converts netrin-1-mediated repulsion to attraction. The Journal of neuroscience : the official journal of the Society for Neuroscience 26, 3192-3205 (2006).

307. Bai, G. & Pfaff, S.L. Protease regulation: the Yin and Yang of neural development and disease. Neuron 72, 9-21 (2011).

308. Sternlicht, M.D. & Werb, Z. How matrix metalloproteinases regulate cell behavior. Annual review of cell and developmental biology 17, 463-516 (2001).

309. Chang, C. & Werb, Z. The many faces of metalloproteases: cell growth, invasion, angiogenesis and metastasis. Trends in cell biology 11, S37-43 (2001).

310. Coleman, H.A., Labrador, J.P., Chance, R.K. & Bashaw, G.J. The Adam family metalloprotease Kuzbanian regulates the cleavage of the roundabout receptor to control axon repulsion at the midline. Development 137, 2417-2426 (2010).

311. Janes, P.W. et al. Adam meets Eph: an ADAM substrate recognition module acts as a molecular switch for ephrin cleavage in trans. Cell 123, 291-304 (2005).

312. Atapattu, L. et al. Antibodies binding the ADAM10 substrate recognition domain inhibit Eph function. Journal of cell science 125, 6084-6093 (2012).

313. Pan, D. & Rubin, G.M. Kuzbanian controls proteolytic processing of Notch and mediates lateral inhibition during Drosophila and vertebrate neurogenesis. Cell 90, 271-280 (1997).

314. Mumm, J.S. et al. A ligand-induced extracellular cleavage regulates gamma-secretase- like proteolytic activation of Notch1. Molecular cell 5, 197-206 (2000).

315. Kageyama, R., Ohtsuka, T., Shimojo, H. & Imayoshi, I. Dynamic Notch signaling in neural progenitor cells and a revised view of lateral inhibition. Nature neuroscience 11, 1247-1251 (2008).

316. Louvi, A. & Artavanis-Tsakonas, S. Notch signalling in vertebrate neural development. Nature reviews. Neuroscience 7, 93-102 (2006).

317. Yang, L., Garbe, D.S. & Bashaw, G.J. A frazzled/DCC-dependent transcriptional switch regulates midline axon guidance. Science 324, 944-947 (2009).

318. Neuhaus-Follini, A. & Bashaw, G.J. The Intracellular Domain of the Frazzled/DCC Receptor Is a Transcription Factor Required for Commissural Axon Guidance. Neuron 87, 751-763 (2015).

319. Tomita, T., Tanaka, S., Morohashi, Y. & Iwatsubo, T. Presenilin-dependent intramembrane cleavage of ephrin-B1. Molecular neurodegeneration 1, 2 (2006).

168

320. Georgakopoulos, A. et al. Metalloproteinase/Presenilin1 processing of ephrinB regulates EphB-induced Src phosphorylation and signaling. The EMBO journal 25, 1242-1252 (2006).

321. Barthet, G. et al. Presenilin mediates neuroprotective functions of ephrinB and brain- derived neurotrophic factor and regulates ligand-induced internalization and metabolism of EphB2 and TrkB receptors. Neurobiology of aging 34, 499-510 (2013).

322. Charoy, C. et al. gdnf activates midline repulsion by Semaphorin3B via NCAM during commissural axon guidance. Neuron 75, 1051-1066 (2012).

323. Shu, T., Valentino, K.M., Seaman, C., Cooper, H.M. & Richards, L.J. Expression of the netrin-1 receptor, deleted in colorectal cancer (DCC), is largely confined to projecting neurons in the developing forebrain. The Journal of comparative neurology 416, 201-212 (2000).

324. Richards, L.J., Koester, S.E., Tuttle, R. & O'Leary, D.D. Directed growth of early cortical axons is influenced by a chemoattractant released from an intermediate target. The Journal of neuroscience : the official journal of the Society for Neuroscience 17, 2445- 2458 (1997).

325. Metin, C., Deleglise, D., Serafini, T., Kennedy, T.E. & Tessier-Lavigne, M. A role for netrin-1 in the guidance of cortical efferents. Development 124, 5063-5074 (1997).

326. Fothergill, T. et al. Netrin-DCC signaling regulates corpus callosum formation through attraction of pioneering axons and by modulating Slit2-mediated repulsion. Cerebral cortex 24, 1138-1151 (2014).

327. Canty, A.J. & Murphy, M. Molecular mechanisms of axon guidance in the developing corticospinal tract. Progress in neurobiology 85, 214-235 (2008).

328. Leyva-Diaz, E. & Lopez-Bendito, G. In and out from the cortex: development of major forebrain connections. Neuroscience 254, 26-44 (2013).

329. Welniarz, Q. et al. Non cell-autonomous role of DCC in the guidance of the corticospinal tract at the midline. Scientific reports 7, 410 (2017).

330. Goll, D.E., Thompson, V.F., Li, H., Wei, W. & Cong, J. The calpain system. Physiological reviews 83, 731-801 (2003).

331. Croall, D.E. & Ersfeld, K. The calpains: modular designs and functional diversity. Genome biology 8, 218 (2007).

332. Sorimachi, H., Hata, S. & Ono, Y. Impact of genetic insights into calpain biology. Journal of biochemistry 150, 23-37 (2011).

169

333. Franco, S.J. & Huttenlocher, A. Regulating cell migration: calpains make the cut. Journal of cell science 118, 3829-3838 (2005).

334. Khorchid, A. & Ikura, M. How calpain is activated by calcium. Nature structural biology 9, 239-241 (2002).

335. Baudry, M. & Bi, X. Calpain-1 and Calpain-2: The Yin and Yang of Synaptic Plasticity and Neurodegeneration. Trends in neurosciences 39, 235-245 (2016).

336. Bevers, M.B. & Neumar, R.W. Mechanistic role of calpains in postischemic neurodegeneration. Journal of cerebral blood flow and metabolism : official journal of the International Society of Cerebral Blood Flow and Metabolism 28, 655-673 (2008).

337. Li, Y., Bondada, V., Joshi, A. & Geddes, J.W. Calpain 1 and Calpastatin expression is developmentally regulated in rat brain. Experimental neurology 220, 316-319 (2009).

338. To, K.C., Church, J. & O'Connor, T.P. Combined activation of calpain and calcineurin during ligand-induced growth cone collapse. Molecular and cellular neurosciences 36, 425-434 (2007).

339. Mingorance-Le Meur, A. & O'Connor, T.P. Neurite consolidation is an active process requiring constant repression of protrusive activity. The EMBO journal 28, 248-260 (2009).

340. Kerstein, P.C., Patel, K.M. & Gomez, T.M. Calpain-Mediated Proteolysis of Talin and FAK Regulates Adhesion Dynamics Necessary for Axon Guidance. The Journal of neuroscience : the official journal of the Society for Neuroscience 37, 1568-1580 (2017).

341. Wang, Y. et al. Defects in the CAPN1 Gene Result in Alterations in Cerebellar Development and Cerebellar Ataxia in Mice and Humans. Cell reports 16, 79-91 (2016).

342. Gil-Parrado, S. et al. Ionomycin-activated calpain triggers apoptosis. A probable role for Bcl-2 family members. The Journal of biological chemistry 277, 27217-27226 (2002).

343. Amini, M. et al. Conditional disruption of calpain in the CNS alters dendrite morphology, impairs LTP, and promotes neuronal survival following injury. The Journal of neuroscience : the official journal of the Society for Neuroscience 33, 5773-5784 (2013).

344. Briz, V. & Baudry, M. Calpains: Master Regulators of Synaptic Plasticity. The Neuroscientist : a review journal bringing neurobiology, neurology and psychiatry (2016).

345. Siman, R., Baudry, M. & Lynch, G. Brain fodrin: substrate for calpain I, an endogenous calcium-activated protease. Proceedings of the National Academy of Sciences of the United States of America 81, 3572-3576 (1984).

170

346. Crocker, S.J. et al. Inhibition of calpains prevents neuronal and behavioral deficits in an MPTP mouse model of Parkinson's disease. The Journal of neuroscience : the official journal of the Society for Neuroscience 23, 4081-4091 (2003).

347. Raynaud, F., Jond-Necand, C., Marcilhac, A., Furst, D. & Benyamin, Y. Calpain 1- gamma filamin interaction in muscle cells: a possible in situ regulation by PKC-alpha. The international journal of biochemistry & cell biology 38, 404-413 (2006).

348. Vosler, P.S., Brennan, C.S. & Chen, J. Calpain-mediated signaling mechanisms in neuronal injury and neurodegeneration. Molecular neurobiology 38, 78-100 (2008).

349. Storr, S.J., Carragher, N.O., Frame, M.C., Parr, T. & Martin, S.G. The calpain system and cancer. Nature reviews. Cancer 11, 364-374 (2011).

350. Ono, Y., Saido, T.C. & Sorimachi, H. Calpain research for drug discovery: challenges and potential. Nature reviews. Drug discovery 15, 854-876 (2016).

351. Suzuki, K., Tsuji, S., Kubota, S., Kimura, Y. & Imahori, K. Limited autolysis of Ca2+- activated neutral protease (CANP) changes its sensitivity to Ca2+ ions. Journal of biochemistry 90, 275-278 (1981).

352. Baki, A., Tompa, P., Alexa, A., Molnar, O. & Friedrich, P. Autolysis parallels activation of mu-calpain. The Biochemical journal 318 ( Pt 3), 897-901 (1996).

353. Suzuki, K., Saido, T.C. & Hirai, S. Modulation of cellular signals by calpain. Annals of the New York Academy of Sciences 674, 218-227 (1992).

354. Saido, T.C., Shibata, M., Takenawa, T., Murofushi, H. & Suzuki, K. Positive regulation of mu-calpain action by polyphosphoinositides. The Journal of biological chemistry 267, 24585-24590 (1992).

355. Shiraha, H., Glading, A., Chou, J., Jia, Z. & Wells, A. Activation of m-calpain (calpain II) by epidermal growth factor is limited by protein kinase A phosphorylation of m- calpain. Molecular and cellular biology 22, 2716-2727 (2002).

356. Glading, A. et al. Epidermal growth factor activates m-calpain (calpain II), at least in part, by extracellular signal-regulated kinase-mediated phosphorylation. Molecular and cellular biology 24, 2499-2512 (2004).

357. Carragher, N.O., Westhoff, M.A., Fincham, V.J., Schaller, M.D. & Frame, M.C. A novel role for FAK as a protease-targeting adaptor protein: regulation by p42 ERK and Src. Current biology : CB 13, 1442-1450 (2003).

358. Liu, Y. et al. Deleting both PHLPP1 and CANP1 rescues impairments in long-term potentiation and learning in both single knockout mice. Learning & memory 23, 399-404 (2016).

171

359. Wang, Y., Briz, V., Chishti, A., Bi, X. & Baudry, M. Distinct roles for mu-calpain and m-calpain in synaptic NMDAR-mediated neuroprotection and extrasynaptic NMDAR- mediated neurodegeneration. The Journal of neuroscience : the official journal of the Society for Neuroscience 33, 18880-18892 (2013).

360. Wendt, A., Thompson, V.F. & Goll, D.E. Interaction of calpastatin with calpain: a review. Biological chemistry 385, 465-472 (2004).

361. De Tullio, R. et al. Changes in intracellular localization of calpastatin during calpain activation. The Biochemical journal 343 Pt 2, 467-472 (1999).

362. Barnoy, S., Zipser, Y., Glaser, T., Grimberg, Y. & Kosower, N.S. Association of calpain (Ca(2+)-dependent thiol protease) with its endogenous inhibitor calpastatin in myoblasts. Journal of cellular biochemistry 74, 522-531 (1999).

363. Serrano, K. & Devine, D.V. Vinculin is proteolyzed by calpain during platelet aggregation: 95 kDa cleavage fragment associates with the platelet cytoskeleton. Cell motility and the cytoskeleton 58, 242-252 (2004).

364. Newcomb, J.K., Pike, B.R., Zhao, X. & Hayes, R.L. Concurrent assessment of calpain and caspase-3 activity by means of western blots of protease-specific spectrin breakdown products. Methods in molecular biology 144, 219-223 (2000).

365. Springer, J.E., Azbill, R.D., Kennedy, S.E., George, J. & Geddes, J.W. Rapid calpain I activation and cytoskeletal protein degradation following traumatic spinal cord injury: attenuation with riluzole pretreatment. Journal of neurochemistry 69, 1592-1600 (1997).

366. McCracken, E., Hunter, A.J., Patel, S., Graham, D.I. & Dewar, D. Calpain activation and cytoskeletal protein breakdown in the corpus callosum of head-injured patients. Journal of neurotrauma 16, 749-761 (1999).

367. Kampfl, A. et al. mu-calpain activation and calpain-mediated cytoskeletal proteolysis following traumatic brain injury. Journal of neurochemistry 67, 1575-1583 (1996).

368. Lebart, M.C. & Benyamin, Y. Calpain involvement in the remodeling of cytoskeletal anchorage complexes. The FEBS journal 273, 3415-3426 (2006).

369. Sedarous, M. et al. Calpains mediate p53 activation and neuronal death evoked by DNA damage. The Journal of biological chemistry 278, 26031-26038 (2003).

370. Qin, Q., Liao, G., Baudry, M. & Bi, X. Role of calpain-mediated p53 truncation in semaphorin 3A-induced axonal growth regulation. Proceedings of the National Academy of Sciences of the United States of America 107, 13883-13887 (2010).

371. Tsai, S.Y., Pokrass, M.J., Klauer, N.R., Nohara, H. & Su, T.P. Sigma-1 receptor regulates Tau phosphorylation and axon extension by shaping p35 turnover via myristic acid.

172

Proceedings of the National Academy of Sciences of the United States of America 112, 6742-6747 (2015).

372. Franco, S.J. et al. Calpain-mediated proteolysis of talin regulates adhesion dynamics. Nature cell biology 6, 977-983 (2004).

373. Bialkowska, K., Saido, T.C. & Fox, J.E. SH3 domain of spectrin participates in the activation of Rac in specialized calpain-induced integrin signaling complexes. Journal of cell science 118, 381-395 (2005).

374. Nedrelow, J.H., Cianci, C.D. & Morrow, J.S. c-Src binds alpha II spectrin's Src homology 3 (SH3) domain and blocks calpain susceptibility by phosphorylating Tyr1176. The Journal of biological chemistry 278, 7735-7741 (2003).

375. Veeranna et al. Calpain mediates calcium-induced activation of the erk1,2 MAPK pathway and cytoskeletal phosphorylation in neurons: relevance to Alzheimer's disease. The American journal of pathology 165, 795-805 (2004).

376. Tompa, P. et al. On the sequential determinants of calpain cleavage. The Journal of biological chemistry 279, 20775-20785 (2004).

377. Kerstein, P.C., Nichol, R.I. & Gomez, T.M. Mechanochemical regulation of growth cone motility. Frontiers in cellular neuroscience 9, 244 (2015).

378. Robles, E., Huttenlocher, A. & Gomez, T.M. Filopodial calcium transients regulate growth cone motility and guidance through local activation of calpain. Neuron 38, 597- 609 (2003).

379. Briz, V. et al. Calpain-2-mediated PTEN degradation contributes to BDNF-induced stimulation of dendritic protein synthesis. The Journal of neuroscience : the official journal of the Society for Neuroscience 33, 4317-4328 (2013).

380. Zadran, S. et al. Brain-derived neurotrophic factor and epidermal growth factor activate neuronal m-calpain via mitogen-activated protein kinase-dependent phosphorylation. The Journal of neuroscience : the official journal of the Society for Neuroscience 30, 1086- 1095 (2010).

381. DeGeer, J. et al. Tyrosine phosphorylation of the Rho guanine nucleotide exchange factor Trio regulates netrin-1/DCC-mediated cortical axon outgrowth. Mol Cell Biol 33, 739- 751 (2013).

382. Endo, M. & Yamashita, T. Inactivation of Ras by p120GAP via focal adhesion kinase dephosphorylation mediates RGMa-induced growth cone collapse. The Journal of neuroscience : the official journal of the Society for Neuroscience 29, 6649-6662 (2009).

173

383. King, P.D., Lubeck, B.A. & Lapinski, P.E. Nonredundant functions for Ras GTPase- activating proteins in tissue homeostasis. Sci Signal 6, re1 (2013).

384. Chengbiao Wu, C.-F.L., and William C. Mobley Nerve Growth Factor Activates Persistent Rap1 Signaling in

Endosomes. J. Neurosci. 21, 5406-5416 (2001).

385. J-H. Duan, Y.W., D. Duarte, M. R. Vasko, G. D. Nicol, and C. M. Hingtgen Ras signaling pathways mediate NGF-induced enhancement of excitability of small-diameter capsaicin-sensitive sensory neurons from wildtype but not Nf1+/− mice. Neurosc. Lett. 496, 70-74 (2011).

386. Kent, C.B. et al. 14-3-3 proteins regulate protein kinase a activity to modulate growth cone turning responses. The Journal of neuroscience : the official journal of the Society for Neuroscience 30, 14059-14067 (2010).

387. Yam, P.T., Langlois, S.D., Morin, S. & Charron, F. Sonic hedgehog guides axons through a noncanonical, Src-family-kinase-dependent signaling pathway. Neuron 62, 349-362 (2009).

388. Serpente, N., Birling, M.C. & Price, J. The regulation of the expression, phosphorylation, and protein associations of pp125FAK during rat brain development. Molecular and cellular neurosciences 7, 391-403 (1996).

389. Hecker, T.P., Ding, Q., Rege, T.A., Hanks, S.K. & Gladson, C.L. Overexpression of FAK promotes Ras activity through the formation of a FAK/p120RasGAP complex in malignant astrocytoma cells. Oncogene 23, 3962-3971 (2004).

390. Holland, S.J. et al. Juxtamembrane tyrosine residues couple the Eph family receptor EphB2/Nuk to specific SH2 domain proteins in neuronal cells. EMBO J 16, 3877-3888 (1997).

391. Hu, K.Q. & Settleman, J. Tandem SH2 binding sites mediate the RasGAP-RhoGAP interaction: a conformational mechanism for SH3 domain regulation. The EMBO journal 16, 473-483 (1997).

392. Kazlauskas, A., Ellis, C., Pawson, T. & Cooper, J.A. Binding of GAP to activated PDGF receptors. Science 247, 1578-1581 (1990).

393. Margolis, B. et al. The tyrosine phosphorylated carboxyterminus of the EGF receptor is a binding site for GAP and PLC-gamma. The EMBO journal 9, 4375-4380 (1990).

174

394. Kerstein, P.C. et al. Mechanosensitive TRPC1 channels promote calpain proteolysis of talin to regulate spinal axon outgrowth. The Journal of neuroscience : the official journal of the Society for Neuroscience 33, 273-285 (2013).

395. Ming, G.L. et al. Adaptation in the chemotactic guidance of nerve growth cones. Nature 417, 411-418 (2002).

396. Rosser, B.G. & Gores, G.J. Cellular in vivo assay of calpain activity using a fluorescent substrate. Application to study of anoxic liver injury. Methods in molecular biology 144, 245-259 (2000).

397. Hancock, M.L., Preitner, N., Quan, J. & Flanagan, J.G. MicroRNA-132 is enriched in developing axons, locally regulates Rasa1 mRNA, and promotes axon extension. The Journal of neuroscience : the official journal of the Society for Neuroscience 34, 66-78 (2014).

398. Elowe, S., Holland, S.J., Kulkarni, S. & Pawson, T. Downregulation of the Ras-mitogen- activated protein kinase pathway by the EphB2 receptor tyrosine kinase is required for ephrin-induced neurite retraction. Molecular and cellular biology 21, 7429-7441 (2001).

399. Skaper, S.D. The neurotrophin family of neurotrophic factors: an overview. Methods in molecular biology 846, 1-12 (2012).

400. van der Geer, P., Henkemeyer, M., Jacks, T. & Pawson, T. Aberrant Ras regulation and reduced p190 tyrosine phosphorylation in cells lacking p120-Gap. Molecular and cellular biology 17, 1840-1847 (1997).

401. Kulkarni, S.V., Gish, G., van der Geer, P., Henkemeyer, M. & Pawson, T. Role of p120 Ras-GAP in directed cell movement. The Journal of cell biology 149, 457-470 (2000).

402. Cacalano, N.A., Sanden, D. & Johnston, J.A. Tyrosine-phosphorylated SOCS-3 inhibits STAT activation but binds to p120 RasGAP and activates Ras. Nature cell biology 3, 460-465 (2001).

403. Shang, X., Moon, S.Y. & Zheng, Y. p200 RhoGAP promotes cell proliferation by mediating cross-talk between Ras and Rho signaling pathways. The Journal of biological chemistry 282, 8801-8811 (2007).

404. Moran, M.F., Polakis, P., McCormick, F., Pawson, T. & Ellis, C. Protein-tyrosine kinases regulate the phosphorylation, protein interactions, subcellular distribution, and activity of p21ras GTPase-activating protein. Molecular and cellular biology 11, 1804-1812 (1991).

405. Nakata, H. & Watanabe, Y. Proliferation and differentiation of PC12 cells were affected by p21ras GTPase activating proteins and its deletion mutant proteins. Biochemical and biophysical research communications 218, 538-543 (1996).

175

406. Leblanc, V., Tocque, B. & Delumeau, I. Ras-GAP controls Rho-mediated cytoskeletal reorganization through its SH3 domain. Molecular and cellular biology 18, 5567-5578 (1998).

407. Tomar, A., Lim, S.T., Lim, Y. & Schlaepfer, D.D. A FAK-p120RasGAP-p190RhoGAP complex regulates polarity in migrating cells. Journal of cell science 122, 1852-1862 (2009).

408. Brouns, M.R., Matheson, S.F. & Settleman, J. p190 RhoGAP is the principal Src substrate in brain and regulates axon outgrowth, guidance and fasciculation. Nature cell biology 3, 361-367 (2001).

409. Yang, X.Y. et al. p120Ras-GAP binds the DLC1 Rho-GAP tumor suppressor protein and inhibits its RhoA GTPase and growth-suppressing activities. Oncogene 28, 1401-1409 (2009).

410. Jaiswal, M. et al. Functional cross-talk between ras and rho pathways: a Ras-specific GTPase-activating protein (p120RasGAP) competitively inhibits the RhoGAP activity of deleted in liver cancer (DLC) tumor suppressor by masking the catalytic arginine finger. The Journal of biological chemistry 289, 6839-6849 (2014).

411. Sabbir, M.G. et al. Identification and characterization of Dlc1 isoforms in the mouse and study of the biological function of a single gene trapped isoform. BMC biology 8, 17 (2010).

412. Braun, A.C. & Olayioye, M.A. Rho regulation: DLC proteins in space and time. Cellular signalling 27, 1643-1651 (2015).

413. Moon, S.Y., Zang, H. & Zheng, Y. Characterization of a brain-specific Rho GTPase- activating protein, p200RhoGAP. The Journal of biological chemistry 278, 4151-4159 (2003).

414. Tang, F. & Kalil, K. Netrin-1 induces axon branching in developing cortical neurons by frequency-dependent calcium signaling pathways. The Journal of neuroscience : the official journal of the Society for Neuroscience 25, 6702-6715 (2005).

415. Lebrand, C. et al. Critical role of Ena/VASP proteins for filopodia formation in neurons and in function downstream of netrin-1. Neuron 42, 37-49 (2004).

416. Bouchard, J.F. et al. Protein kinase A activation promotes plasma membrane insertion of DCC from an intracellular pool: A novel mechanism regulating commissural axon extension. The Journal of neuroscience : the official journal of the Society for Neuroscience 24, 3040-3050 (2004).

417. Dickson, B.J. & Zou, Y. Navigating intermediate targets: the nervous system midline. Cold Spring Harbor perspectives in biology 2, a002055 (2010).

176

418. Bendriem, R.M. & Ross, M.E. Wiring the Human Brain: A User's Handbook. Neuron 95, 482-485 (2017).

419. Fazeli, A. et al. Phenotype of mice lacking functional Deleted in colorectal cancer (Dcc) gene. Nature 386, 796-804 (1997).

420. Serafini, T. et al. Netrin-1 is required for commissural axon guidance in the developing vertebrate nervous system. Cell 87, 1001-1014 (1996).

421. Jamuar, S.S. et al. Biallelic mutations in human DCC cause developmental split-brain syndrome. Nature genetics 49, 606-612 (2017).

422. Depienne, C. et al. A novel DCC mutation and genetic heterogeneity in congenital mirror movements. Neurology 76, 260-264 (2011).

423. Bierhals, T. et al. Novel DCC variants in congenital mirror movements and evaluation of disease-associated missense variants. European journal of medical genetics (2018).

424. Gan-Or, Z. et al. Mutations in CAPN1 Cause Autosomal-Recessive Hereditary Spastic Paraplegia. American journal of human genetics 98, 1038-1046 (2016).

425. Tompa, P., Baki, A., Schad, E. & Friedrich, P. The calpain cascade. Mu-calpain activates m-calpain. The Journal of biological chemistry 271, 33161-33164 (1996).

426. Gomez, T.M. & Spitzer, N.C. In vivo regulation of axon extension and pathfinding by growth-cone calcium transients. Nature 397, 350-355 (1999).

427. Nishiyama, M., von Schimmelmann, M.J., Togashi, K., Findley, W.M. & Hong, K. Membrane potential shifts caused by diffusible guidance signals direct growth-cone turning. Nature neuroscience 11, 762-771 (2008).

428. Nishiyama, M. et al. Cyclic AMP/GMP-dependent modulation of Ca2+ channels sets the polarity of nerve growth-cone turning. Nature 423, 990-995 (2003).

429. Nicol, X., Hong, K.P. & Spitzer, N.C. Spatial and temporal second messenger codes for growth cone turning. Proceedings of the National Academy of Sciences of the United States of America 108, 13776-13781 (2011).

430. Hong, K., Nishiyama, M., Henley, J., Tessier-Lavigne, M. & Poo, M. Calcium signalling in the guidance of nerve growth by netrin-1. Nature 403, 93-98 (2000).

431. De Strooper, B. & Annaert, W. Novel research horizons for presenilins and gamma- secretases in cell biology and disease. Annual review of cell and developmental biology 26, 235-260 (2010).

177

432. Ardito, F., Giuliani, M., Perrone, D., Troiano, G. & Lo Muzio, L. The crucial role of protein phosphorylation in cell signaling and its use as targeted therapy (Review). International journal of molecular medicine 40, 271-280 (2017).

433. Nishi, H., Shaytan, A. & Panchenko, A.R. Physicochemical mechanisms of protein regulation by phosphorylation. Frontiers in genetics 5, 270 (2014).

434. Saito, T. et al. Developmental regulation of the proteolysis of the p35 cyclin-dependent kinase 5 activator by phosphorylation. The Journal of neuroscience : the official journal of the Society for Neuroscience 23, 1189-1197 (2003).

435. Chen, M. & Stracher, A. In situ phosphorylation of platelet actin-binding protein by cAMP-dependent protein kinase stabilizes it against proteolysis by calpain. The Journal of biological chemistry 264, 14282-14289 (1989).

436. Wang, H. et al. PKA-mediated protein phosphorylation protects ezrin from calpain I cleavage. Biochemical and biophysical research communications 333, 496-501 (2005).

437. Walter, J., Schindzielorz, A., Grunberg, J. & Haass, C. Phosphorylation of presenilin-2 regulates its cleavage by caspases and retards progression of apoptosis. Proceedings of the National Academy of Sciences of the United States of America 96, 1391-1396 (1999).

438. Spizz, G. & Blackshear, P.J. Protein kinase C-mediated phosphorylation of the myristoylated alanine-rich C-kinase substrate protects it from specific proteolytic cleavage. The Journal of biological chemistry 271, 553-562 (1996).

439. Price, M.A. & Kalderon, D. Proteolysis of the Hedgehog signaling effector Cubitus interruptus requires phosphorylation by Glycogen Synthase Kinase 3 and Casein Kinase 1. Cell 108, 823-835 (2002).

440. Mailand, N. & Diffley, J.F. CDKs promote DNA replication origin licensing in human cells by protecting Cdc6 from APC/C-dependent proteolysis. Cell 122, 915-926 (2005).

441. Greenwood, J.A., Troncoso, J.C., Costello, A.C. & Johnson, G.V. Phosphorylation modulates calpain-mediated proteolysis and calmodulin binding of the 200-kDa and 160- kDa neurofilament proteins. Journal of neurochemistry 61, 191-199 (1993).

442. Morin, V. et al. Phosphorylation protects sperm-specific histones H1 and H2B from proteolysis after fertilization. Journal of cellular biochemistry 76, 173-180 (1999).

443. Zhang, Z., Lawrence, J. & Stracher, A. Phosphorylation of platelet actin binding protein protects against proteolysis by calcium dependent sulfhydryl protease. Biochemical and biophysical research communications 151, 355-360 (1988).

444. Goldstein, M.E., Sternberger, N.H. & Sternberger, L.A. Phosphorylation protects neurofilaments against proteolysis. Journal of neuroimmunology 14, 149-160 (1987).

178

445. Lange, C.A., Shen, T. & Horwitz, K.B. Phosphorylation of human progesterone receptors at serine-294 by mitogen-activated protein kinase signals their degradation by the 26S proteasome. Proceedings of the National Academy of Sciences of the United States of America 97, 1032-1037 (2000).

446. Lee, M.S. et al. Neurotoxicity induces cleavage of p35 to p25 by calpain. Nature 405, 360-364 (2000).

447. Dou, H. et al. Structural basis for autoinhibition and phosphorylation-dependent activation of c-Cbl. Nature structural & molecular biology 19, 184-192 (2012).

448. Zhou, A., Webb, G., Zhu, X. & Steiner, D.F. Proteolytic processing in the secretory pathway. The Journal of biological chemistry 274, 20745-20748 (1999).

449. Wei, Z., Yan, J., Lu, Q., Pan, L. & Zhang, M. Cargo recognition mechanism of myosin X revealed by the structure of its tail MyTH4-FERM tandem in complex with the DCC P3 domain. Proceedings of the National Academy of Sciences of the United States of America 108, 3572-3577 (2011).

450. Chan, S.L. & Mattson, M.P. Caspase and calpain substrates: roles in synaptic plasticity and cell death. Journal of neuroscience research 58, 167-190 (1999).

451. Sareen, D., Darjatmoko, S.R., Albert, D.M. & Polans, A.S. Mitochondria, calcium, and calpain are key mediators of resveratrol-induced apoptosis in breast cancer. Molecular pharmacology 72, 1466-1475 (2007).

452. Negulescu, A.M. & Mehlen, P. Dependence receptors - the dark side awakens. The FEBS journal (2018).

453. Kennedy, T.E., Wang, H., Marshall, W. & Tessier-Lavigne, M. Axon guidance by diffusible chemoattractants: a gradient of netrin protein in the developing spinal cord. J Neurosci 26, 8866-8874 (2006).

454. Hayashi, K., Yonemura, S., Matsui, T. & Tsukita, S. Immunofluorescence detection of ezrin/radixin/moesin (ERM) proteins with their carboxyl-terminal threonine phosphorylated in cultured cells and tissues. J Cell Sci 112 ( Pt 8), 1149-1158 (1999).

179